diff --git a/CHANGELOG b/CHANGELOG new file mode 100644 --- /dev/null +++ b/CHANGELOG @@ -0,0 +1,8 @@ +# v1.5.0.1 _(2020-02-24)_ +- Added `CHANGELOG` to its own file. +- Included `CHANGELOG` and `README.pro` in the source distribution. +- Added an optic for fetching the name of a `Tag`. + +# v1.5.0.0 _(2020-02-23)_ +- Added tests for GHC 8.4.4 through 8.6.1. +- Initial stable release. diff --git a/LICENSE b/LICENSE new file mode 100644 --- /dev/null +++ b/LICENSE @@ -0,0 +1,373 @@ +Mozilla Public License Version 2.0 +================================== + +1. Definitions +-------------- + +1.1. "Contributor" + means each individual or legal entity that creates, contributes to + the creation of, or owns Covered Software. + +1.2. "Contributor Version" + means the combination of the Contributions of others (if any) used + by a Contributor and that particular Contributor's Contribution. + +1.3. "Contribution" + means Covered Software of a particular Contributor. + +1.4. "Covered Software" + means Source Code Form to which the initial Contributor has attached + the notice in Exhibit A, the Executable Form of such Source Code + Form, and Modifications of such Source Code Form, in each case + including portions thereof. + +1.5. "Incompatible With Secondary Licenses" + means + + (a) that the initial Contributor has attached the notice described + in Exhibit B to the Covered Software; or + + (b) that the Covered Software was made available under the terms of + version 1.1 or earlier of the License, but not also under the + terms of a Secondary License. + +1.6. "Executable Form" + means any form of the work other than Source Code Form. + +1.7. "Larger Work" + means a work that combines Covered Software with other material, in + a separate file or files, that is not Covered Software. + +1.8. "License" + means this document. + +1.9. "Licensable" + means having the right to grant, to the maximum extent possible, + whether at the time of the initial grant or subsequently, any and + all of the rights conveyed by this License. + +1.10. "Modifications" + means any of the following: + + (a) any file in Source Code Form that results from an addition to, + deletion from, or modification of the contents of Covered + Software; or + + (b) any new file in Source Code Form that contains any Covered + Software. + +1.11. "Patent Claims" of a Contributor + means any patent claim(s), including without limitation, method, + process, and apparatus claims, in any patent Licensable by such + Contributor that would be infringed, but for the grant of the + License, by the making, using, selling, offering for sale, having + made, import, or transfer of either its Contributions or its + Contributor Version. + +1.12. "Secondary License" + means either the GNU General Public License, Version 2.0, the GNU + Lesser General Public License, Version 2.1, the GNU Affero General + Public License, Version 3.0, or any later versions of those + licenses. + +1.13. "Source Code Form" + means the form of the work preferred for making modifications. + +1.14. "You" (or "Your") + means an individual or a legal entity exercising rights under this + License. For legal entities, "You" includes any entity that + controls, is controlled by, or is under common control with You. For + purposes of this definition, "control" means (a) the power, direct + or indirect, to cause the direction or management of such entity, + whether by contract or otherwise, or (b) ownership of more than + fifty percent (50%) of the outstanding shares or beneficial + ownership of such entity. + +2. License Grants and Conditions +-------------------------------- + +2.1. Grants + +Each Contributor hereby grants You a world-wide, royalty-free, +non-exclusive license: + +(a) under intellectual property rights (other than patent or trademark) + Licensable by such Contributor to use, reproduce, make available, + modify, display, perform, distribute, and otherwise exploit its + Contributions, either on an unmodified basis, with Modifications, or + as part of a Larger Work; and + +(b) under Patent Claims of such Contributor to make, use, sell, offer + for sale, have made, import, and otherwise transfer either its + Contributions or its Contributor Version. + +2.2. Effective Date + +The licenses granted in Section 2.1 with respect to any Contribution +become effective for each Contribution on the date the Contributor first +distributes such Contribution. + +2.3. Limitations on Grant Scope + +The licenses granted in this Section 2 are the only rights granted under +this License. No additional rights or licenses will be implied from the +distribution or licensing of Covered Software under this License. +Notwithstanding Section 2.1(b) above, no patent license is granted by a +Contributor: + +(a) for any code that a Contributor has removed from Covered Software; + or + +(b) for infringements caused by: (i) Your and any other third party's + modifications of Covered Software, or (ii) the combination of its + Contributions with other software (except as part of its Contributor + Version); or + +(c) under Patent Claims infringed by Covered Software in the absence of + its Contributions. + +This License does not grant any rights in the trademarks, service marks, +or logos of any Contributor (except as may be necessary to comply with +the notice requirements in Section 3.4). + +2.4. Subsequent Licenses + +No Contributor makes additional grants as a result of Your choice to +distribute the Covered Software under a subsequent version of this +License (see Section 10.2) or under the terms of a Secondary License (if +permitted under the terms of Section 3.3). + +2.5. Representation + +Each Contributor represents that the Contributor believes its +Contributions are its original creation(s) or it has sufficient rights +to grant the rights to its Contributions conveyed by this License. + +2.6. Fair Use + +This License is not intended to limit any rights You have under +applicable copyright doctrines of fair use, fair dealing, or other +equivalents. + +2.7. Conditions + +Sections 3.1, 3.2, 3.3, and 3.4 are conditions of the licenses granted +in Section 2.1. + +3. Responsibilities +------------------- + +3.1. Distribution of Source Form + +All distribution of Covered Software in Source Code Form, including any +Modifications that You create or to which You contribute, must be under +the terms of this License. You must inform recipients that the Source +Code Form of the Covered Software is governed by the terms of this +License, and how they can obtain a copy of this License. You may not +attempt to alter or restrict the recipients' rights in the Source Code +Form. + +3.2. Distribution of Executable Form + +If You distribute Covered Software in Executable Form then: + +(a) such Covered Software must also be made available in Source Code + Form, as described in Section 3.1, and You must inform recipients of + the Executable Form how they can obtain a copy of such Source Code + Form by reasonable means in a timely manner, at a charge no more + than the cost of distribution to the recipient; and + +(b) You may distribute such Executable Form under the terms of this + License, or sublicense it under different terms, provided that the + license for the Executable Form does not attempt to limit or alter + the recipients' rights in the Source Code Form under this License. + +3.3. Distribution of a Larger Work + +You may create and distribute a Larger Work under terms of Your choice, +provided that You also comply with the requirements of this License for +the Covered Software. If the Larger Work is a combination of Covered +Software with a work governed by one or more Secondary Licenses, and the +Covered Software is not Incompatible With Secondary Licenses, this +License permits You to additionally distribute such Covered Software +under the terms of such Secondary License(s), so that the recipient of +the Larger Work may, at their option, further distribute the Covered +Software under the terms of either this License or such Secondary +License(s). + +3.4. Notices + +You may not remove or alter the substance of any license notices +(including copyright notices, patent notices, disclaimers of warranty, +or limitations of liability) contained within the Source Code Form of +the Covered Software, except that You may alter any license notices to +the extent required to remedy known factual inaccuracies. + +3.5. Application of Additional Terms + +You may choose to offer, and to charge a fee for, warranty, support, +indemnity or liability obligations to one or more recipients of Covered +Software. However, You may do so only on Your own behalf, and not on +behalf of any Contributor. You must make it absolutely clear that any +such warranty, support, indemnity, or liability obligation is offered by +You alone, and You hereby agree to indemnify every Contributor for any +liability incurred by such Contributor as a result of warranty, support, +indemnity or liability terms You offer. You may include additional +disclaimers of warranty and limitations of liability specific to any +jurisdiction. + +4. Inability to Comply Due to Statute or Regulation +--------------------------------------------------- + +If it is impossible for You to comply with any of the terms of this +License with respect to some or all of the Covered Software due to +statute, judicial order, or regulation then You must: (a) comply with +the terms of this License to the maximum extent possible; and (b) +describe the limitations and the code they affect. Such description must +be placed in a text file included with all distributions of the Covered +Software under this License. Except to the extent prohibited by statute +or regulation, such description must be sufficiently detailed for a +recipient of ordinary skill to be able to understand it. + +5. Termination +-------------- + +5.1. The rights granted under this License will terminate automatically +if You fail to comply with any of its terms. However, if You become +compliant, then the rights granted under this License from a particular +Contributor are reinstated (a) provisionally, unless and until such +Contributor explicitly and finally terminates Your grants, and (b) on an +ongoing basis, if such Contributor fails to notify You of the +non-compliance by some reasonable means prior to 60 days after You have +come back into compliance. Moreover, Your grants from a particular +Contributor are reinstated on an ongoing basis if such Contributor +notifies You of the non-compliance by some reasonable means, this is the +first time You have received notice of non-compliance with this License +from such Contributor, and You become compliant prior to 30 days after +Your receipt of the notice. + +5.2. If You initiate litigation against any entity by asserting a patent +infringement claim (excluding declaratory judgment actions, +counter-claims, and cross-claims) alleging that a Contributor Version +directly or indirectly infringes any patent, then the rights granted to +You by any and all Contributors for the Covered Software under Section +2.1 of this License shall terminate. + +5.3. In the event of termination under Sections 5.1 or 5.2 above, all +end user license agreements (excluding distributors and resellers) which +have been validly granted by You or Your distributors under this License +prior to termination shall survive termination. + +************************************************************************ +* * +* 6. Disclaimer of Warranty * +* ------------------------- * +* * +* Covered Software is provided under this License on an "as is" * +* basis, without warranty of any kind, either expressed, implied, or * +* statutory, including, without limitation, warranties that the * +* Covered Software is free of defects, merchantable, fit for a * +* particular purpose or non-infringing. The entire risk as to the * +* quality and performance of the Covered Software is with You. * +* Should any Covered Software prove defective in any respect, You * +* (not any Contributor) assume the cost of any necessary servicing, * +* repair, or correction. This disclaimer of warranty constitutes an * +* essential part of this License. No use of any Covered Software is * +* authorized under this License except under this disclaimer. * +* * +************************************************************************ + +************************************************************************ +* * +* 7. Limitation of Liability * +* -------------------------- * +* * +* Under no circumstances and under no legal theory, whether tort * +* (including negligence), contract, or otherwise, shall any * +* Contributor, or anyone who distributes Covered Software as * +* permitted above, be liable to You for any direct, indirect, * +* special, incidental, or consequential damages of any character * +* including, without limitation, damages for lost profits, loss of * +* goodwill, work stoppage, computer failure or malfunction, or any * +* and all other commercial damages or losses, even if such party * +* shall have been informed of the possibility of such damages. This * +* limitation of liability shall not apply to liability for death or * +* personal injury resulting from such party's negligence to the * +* extent applicable law prohibits such limitation. Some * +* jurisdictions do not allow the exclusion or limitation of * +* incidental or consequential damages, so this exclusion and * +* limitation may not apply to You. * +* * +************************************************************************ + +8. Litigation +------------- + +Any litigation relating to this License may be brought only in the +courts of a jurisdiction where the defendant maintains its principal +place of business and such litigation shall be governed by laws of that +jurisdiction, without reference to its conflict-of-law provisions. +Nothing in this Section shall prevent a party's ability to bring +cross-claims or counter-claims. + +9. Miscellaneous +---------------- + +This License represents the complete agreement concerning the subject +matter hereof. If any provision of this License is held to be +unenforceable, such provision shall be reformed only to the extent +necessary to make it enforceable. Any law or regulation which provides +that the language of a contract shall be construed against the drafter +shall not be used to construe this License against a Contributor. + +10. Versions of the License +--------------------------- + +10.1. New Versions + +Mozilla Foundation is the license steward. Except as provided in Section +10.3, no one other than the license steward has the right to modify or +publish new versions of this License. Each version will be given a +distinguishing version number. + +10.2. Effect of New Versions + +You may distribute the Covered Software under the terms of the version +of the License under which You originally received the Covered Software, +or under the terms of any subsequent version published by the license +steward. + +10.3. Modified Versions + +If you create software not governed by this License, and you want to +create a new license for such software, you may create and use a +modified version of this License if you rename the license and remove +any references to the name of the license steward (except to note that +such modified license differs from this License). + +10.4. Distributing Source Code Form that is Incompatible With Secondary +Licenses + +If You choose to distribute Source Code Form that is Incompatible With +Secondary Licenses under the terms of this version of the License, the +notice described in Exhibit B of this License must be attached. + +Exhibit A - Source Code Form License Notice +------------------------------------------- + + This Source Code Form is subject to the terms of the Mozilla Public + License, v. 2.0. If a copy of the MPL was not distributed with this + file, You can obtain one at http://mozilla.org/MPL/2.0/. + +If it is not possible or desirable to put the notice in a particular +file, then You may include the notice in a location (such as a LICENSE +file in a relevant directory) where a recipient would be likely to look +for such a notice. + +You may add additional accurate notices of copyright ownership. + +Exhibit B - "Incompatible With Secondary Licenses" Notice +--------------------------------------------------------- + + This Source Code Form is "Incompatible With Secondary Licenses", as + defined by the Mozilla Public License, v. 2.0. diff --git a/README.pro b/README.pro new file mode 100644 --- /dev/null +++ b/README.pro @@ -0,0 +1,33 @@ +title: Prosidy README +--- + +Prosidy is a small language for writing documents. + +Like #link[url='https://daringfireball.net/projects/markdown/']{Markdown}, +Prosidy's syntax is lightweight; it doesn't get in the way of your text. + +Like #link[url='https://www.w3.org/XML/']{XML}, +Prosidy is extensible: it doesn't make any assumptions about your content. +You'll never have to fight to make your data fit a structure that wasn't designed for it. + +#=src[lang='prosidy']:end +recipe: A simple roux +--- +In a #ware{medium saucepan}, +heat #ingredient[amount='25', unit='g']{oil} +until dropping a pinch of flour into the oil causes it to bubble. + +#ware{Whisk} #ingredient[amount='25', unit='g']{flour} +into the oil until the roux is the desired color. +#:end + +All of the code here is under #b{heavy} development; +be careful before using it for anything critical! + +That said, feedback is more than welcome! +Reach me at #link[url='mailto:alex@fldcr.com']{alex@fldcr.com}. + +#-h{Developing Prosidy} +This project sets up an environment using direnv and Nix. +We'd recommend you have both installed before building, +although it may work without both. diff --git a/data/golden/comment/input.pro b/data/golden/comment/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/comment/input.pro @@ -0,0 +1,127 @@ +case: Comments + +## Empty line in header + ## Empty line after spaces in header + +prop ## After a property +empty = ## After an empty setting +nonempty = yes ## After a non-empty setting + +--- + +## Empty line in the body + ## Empty line after spaces in the body + + +#-tag ## Immediately after an empty block-tag + +## On the line preceeding an empty block tag +#-tag + +#-tag +## On the line proceeding an empty block tag + + +#-tag[prop, setting=''] ## Immediately after an empty block-tag with metadata + +## On the line preceeding an empty block tag with metadata +#-tag[prop, setting=''] + +#-tag[prop, setting=''] +## On the line proceeding an empty block tag with metadata + +#-tag[ ## After the metadata open bracket +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +] ## After a metadata close bracket + +#-tag[ ## After the metadata open bracket +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +]: ## After the block open colon + ## Within a block + Inner contents + + Another paragraph + ## Within a block +#: ## After a named block close + +#-tag[ ## After the metadata open bracket +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +]:end ## After the block open colon + ## Within a block + Inner contents + + Another paragraph + ## Within a block +#:end ## After a named block close + +#-tag[ ## After the metadata open bracket +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +]{ ## After the inline brace + +## Within the brace + +Inner contents + +## Within the brace + +} ## After the inline brace + +## Before a paragraph +Blah blah ## After a line in a paragraph +blah blah. ## After a line in a paragraph +#inline[ ## After a dangling inline meta open +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +] ## After the inline meta close +#inline[ ## After a dangling inline meta open +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +]{ ## Within the brace + ## Within the brace +look ma ## After a paragraph in a brace + ## Within the brace + ## Within the brace +} ## After a brace closes +#inline[ ## After a dangling inline meta open +prop ## After a property in the metadata +, ## After a metadata separator +setting ## After a setting key in the metadata += ## After the association operator +"foo" ## After a quoted value +]{ ## Within the brace + ## Within the brace +no hands ## After a paragraph in a brace + ## Within the brace + ## Within the brace +} content ## After a brace closes with trailing content +## After a paragraph + + +#=src:end ## After a literal tag's opening line + ## This is _not_ a comment; it's part of the tag around it +#:end ## After a literal tag's closing line + +## As the last line of a document, without a trailing newline +## This is a direct remediation of a bug I found diff --git a/data/golden/comment/output.json b/data/golden/comment/output.json new file mode 100644 --- /dev/null +++ b/data/golden/comment/output.json @@ -0,0 +1,312 @@ +{ + "content": [ + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": {}, + "properties": {} + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": {}, + "properties": {} + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": {}, + "properties": {} + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": { + "setting": "" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": { + "setting": "" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": { + "setting": "" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [], + "name": "tag", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [ + { + "value": [ + { + "value": "Inner contents", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Another paragraph", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "name": "tag", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [ + { + "value": [ + { + "value": "Inner contents", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Another paragraph", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "name": "tag", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": { + "content": [ + { + "value": [ + { + "value": "Inner contents", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "name": "tag", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "block", + "type": "tag" + }, + { + "value": [ + { + "value": "Blah blah", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "blah blah.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "content": [], + "name": "inline", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "inline", + "type": "tag" + }, + { + "type": "break" + }, + { + "value": { + "content": [ + { + "value": "look ma", + "type": "text" + } + ], + "name": "inline", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "inline", + "type": "tag" + }, + { + "type": "break" + }, + { + "value": { + "content": [ + { + "value": "no hands", + "type": "text" + } + ], + "name": "inline", + "metadata": { + "settings": { + "setting": "foo" + }, + "properties": { + "prop": true + } + } + }, + "subtype": "inline", + "type": "tag" + }, + { + "type": "break" + }, + { + "value": "content", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": { + "content": " ## This is _not_ a comment; it's part of the tag around it", + "name": "src", + "metadata": { + "settings": {}, + "properties": {} + } + }, + "subtype": "literal", + "type": "tag" + } + ], + "metadata": { + "settings": { + "empty": "", + "case": "Comments", + "nonempty": "yes" + }, + "properties": { + "prop": true + } + } +} diff --git a/data/golden/empty-document/input.pro b/data/golden/empty-document/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/empty-document/input.pro @@ -0,0 +1,1 @@ +--- diff --git a/data/golden/empty-document/output.json b/data/golden/empty-document/output.json new file mode 100644 --- /dev/null +++ b/data/golden/empty-document/output.json @@ -0,0 +1,7 @@ +{ + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] +} diff --git a/data/golden/gamut/input.pro b/data/golden/gamut/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/gamut/input.pro @@ -0,0 +1,164 @@ +## This is an example document which enumerates many valid syntaxes. It is +## probably not exhaustive, but should serve as a good document to implement +## against. + +## Properties +prop +prop-with-comment ## this has a trailing comment + prop-with-leading-spaces ## arbitrary leading spaces are permitted + +## Settings +set-a1:42 +set-a2=42 +set-a3: 42 +set-a4= 42 +set-a5 :42 +set-a6 =42 +set-a7 = 42 +set-a8 : 42 + + set-a9 = 42 ## arbitrary leading spaces are permitted + +## Settings with trailing comments +set-b1:42 ## this has a trailing comment +set-b2=42 ## this has a trailing comment +set-b3: 42 ## this has a trailing comment +set-b4= 42 ## this has a trailing comment +set-b5 :42 ## this has a trailing comment +set-b6 =42 ## this has a trailing comment +set-b7 = 42 ## this has a trailing comment +set-b8 : 42 ## this has a trailing comment + +## Settings with escaped values +set-c1 = foo \# bar +set-c2 = \#\# not a comment ## is a comment + +## Settings with null values +set-d1 = +set-d2 : +set-d3 = ## Just a comment, not a value +set-d4 : ## Just a comment, not a value +--- + +Paragraph with only one line. + +Paragraph trailed by comment. ## This is the comment + +Paragraph split +over multiple lines. + + Paragraph split ## with an indent! +over multiple lines ## Another sneaky comment +with a comment trailing some. + +#tag{This} paragraph starts with a simple tag. + +This paragraph ends with a simple #tag{tag.} + +This #tag{paragraph} has a few #tag{tags}. + +#a{}b#c{}#d{} + +This paragraph has a number of inline tag styles. +#tag ## no meta, no children +#tag[] ## no meta, no children +#tag[]{} ## no meta, no children +#tag[prop] ## 1 prop, no children +#tag[prop1,prop2] ## 2 props, no children +#tag[set = '42'] ## 1 setting (equal, single quote), no children +#tag[set = "42"] ## 1 setting (equal, double quote), no children +#tag[set: '42'] ## 1 setting (colon, single quote), no children +#tag[set: "42"] ## 1 setting (colon, double quote), no children +#tag[prop, foo: 'bar', baz = "quz"] ## complicated, no children +#tag{content} ## no meta, children +#tag[]{content} ## no meta, children +#tag{#tag{ok}} ## no meta, children containing tags +#tag[prop, foo: 'bar', baz = "quz"]{#tag{ok}} ## ok wow +#tag[ +] ## Meta blocks can span multiple lines +#tag[ ## Comments can go here + ## Because items in the metadata block are relatively structured + ## compared to the rest of the document, you can go wild with + ## how properties are laid out + + prop1, prop2 ## Comments can go here +, set ## Comments can go here += '42', prop3 ## Comments can go here +]{This one is pathological} + +This paragraph has some tags that are awfuly close to one another. +#tag{}text#tag{}text#tag#tag{}. + +#tag{This is a paragraph in an #kind{inline} tag.} + +#tag{This is a paragraph in an #kind{inline} tag. + It spans many lines. + Content may follow.} This follows! + +#tag{This is a paragraph in an #kind{inline} tag. ## Comments can go here + It spans many lines . ## Comments can go here + Content may follow.} This follows! ## Comments can go here + +#-tag{This is a paragraph in a #kind{block} tag. ## Comments can go here + Content may not follow.} ## Comments can go here + +#-block: ## Comments can go here + This is the other kind of block tag. + It can have #italic{nested} block children. + + #-block: ## Empty blocks are fine too + #: + + #-block[prop]:label ## Comments can go here + Block starts/ends can have aribtrary labels attached. + #:label ## Comments can go here + + #-block[set='42']:label ## Comments can go here + Block starts/ends can have aribtrary labels attached. + #:label ## Comments can go here + + #-block[prop1, set: '42', prop2]:label ## Comments can go here + Block starts/ends can have aribtrary labels attached. + #:label ## Comments can go here +#: ## Comments can go here + +#-block[ ## Comments can go here + ## Because items in the metadata block are relatively structured + ## compared to the rest of the document, you can go wild with + ## how properties are laid out + + prop1, prop2 ## Comments can go here +, set ## Comments can go here += '42', prop3 ## Comments can go here +]:end ## Comments can go here + This is another pathological tag, just a block this time. + #:end + +## This is a literal tag +## The label is once again optional like block tags, +## but a label is _very_ helpful for quoting prosidy +#=lit:eof ## Comments can go here, but not on the next line +#-block: ## this is treated literally! + This is the other kind of block tag. + It can have #italic{nested} block children. + + #-block: ## this is treated literally! + #: + + #-block[prop]:label ## this is treated literally! + Block starts/ends can have aribtrary labels attached. + #:label ## this is treated literally! + + #-block[set='42']:label ## this is treated literally! + Block starts/ends can have aribtrary labels attached. + #:label ## this is treated literally! + + #-block[prop1, set: '42', prop2]:label ## this is treated literally! + Block starts/ends can have aribtrary labels attached. + #:label ## this is treated literally! +#: ## this is treated literally! +#:eof ## Comments can go here again + +Some unicode escapes: + "\u1F638" is a smiling cat face, + and "\uB" is a vertical tab. diff --git a/data/golden/gamut/output.json b/data/golden/gamut/output.json new file mode 100644 --- /dev/null +++ b/data/golden/gamut/output.json @@ -0,0 +1,1108 @@ +{ + "metadata": { + "properties": { + "prop": true, + "prop-with-comment": true, + "prop-with-leading-spaces": true + }, + "settings": { + "set-a1": "42", + "set-a2": "42", + "set-a3": "42", + "set-a4": "42", + "set-a5": "42", + "set-a6": "42", + "set-a7": "42", + "set-a8": "42", + "set-a9": "42", + "set-b1": "42", + "set-b2": "42", + "set-b3": "42", + "set-b4": "42", + "set-b5": "42", + "set-b6": "42", + "set-b7": "42", + "set-b8": "42", + "set-c1": "foo # bar", + "set-c2": "## not a comment", + "set-d1": "", + "set-d2": "", + "set-d3": "", + "set-d4": "" + } + }, + "content": [ + { + "value": [ + { + "value": "Paragraph with only one line.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Paragraph trailed by comment.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Paragraph split", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "over multiple lines.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Paragraph split", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "over multiple lines", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "with a comment trailing some.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "This", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "paragraph starts with a simple tag.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This paragraph ends with a simple", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "tag.", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "paragraph", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "has a few", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "tags", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": ".", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": { + "name": "a", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": "b", + "type": "text" + }, + { + "value": { + "name": "c", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": { + "name": "d", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This paragraph has a number of inline tag styles.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": { + "prop": true + }, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": { + "prop1": true, + "prop2": true + }, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": { + "set": "42" + } + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": { + "set": "42" + } + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": { + "set": "42" + } + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": { + "set": "42" + } + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": { + "prop": true + }, + "settings": { + "baz": "quz", + "foo": "bar" + } + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "content", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "content", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "ok", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": { + "prop": true + }, + "settings": { + "baz": "quz", + "foo": "bar" + } + }, + "content": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "ok", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": { + "prop1": true, + "prop2": true, + "prop3": true + }, + "settings": { + "set": "42" + } + }, + "content": [ + { + "value": "This one is pathological", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This paragraph has some tags that are awfuly close to one another.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": "text", + "type": "text" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": "text", + "type": "text" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "inline" + }, + { + "value": ".", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "This is a paragraph in an", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "kind", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "inline", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "tag.", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "This is a paragraph in an", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "kind", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "inline", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "tag.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "It spans many lines.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "Content may follow.", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "This follows!", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "This is a paragraph in an", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "kind", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "inline", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "tag.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "It spans many lines .", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "Content may follow.", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "This follows!", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": { + "name": "tag", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": [ + { + "value": "This is a paragraph in a", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "kind", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "block", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "tag.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "Content may not follow.", + "type": "text" + } + ], + "type": "paragraph" + } + ] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": [ + { + "value": "This is the other kind of block tag.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "It can have", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "name": "italic", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [ + { + "value": "nested", + "type": "text" + } + ] + }, + "type": "tag", + "subtype": "inline" + }, + { + "type": "break" + }, + { + "value": "block children.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": [] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": { + "prop": true + }, + "settings": {} + }, + "content": [ + { + "value": [ + { + "value": "Block starts/ends can have aribtrary labels attached.", + "type": "text" + } + ], + "type": "paragraph" + } + ] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": {}, + "settings": { + "set": "42" + } + }, + "content": [ + { + "value": [ + { + "value": "Block starts/ends can have aribtrary labels attached.", + "type": "text" + } + ], + "type": "paragraph" + } + ] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": { + "prop1": true, + "prop2": true + }, + "settings": { + "set": "42" + } + }, + "content": [ + { + "value": [ + { + "value": "Block starts/ends can have aribtrary labels attached.", + "type": "text" + } + ], + "type": "paragraph" + } + ] + }, + "type": "tag", + "subtype": "block" + } + ] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "block", + "metadata": { + "properties": { + "prop1": true, + "prop2": true, + "prop3": true + }, + "settings": { + "set": "42" + } + }, + "content": [ + { + "value": [ + { + "value": "This is another pathological tag, just a block this time.", + "type": "text" + } + ], + "type": "paragraph" + } + ] + }, + "type": "tag", + "subtype": "block" + }, + { + "value": { + "name": "lit", + "metadata": { + "properties": {}, + "settings": {} + }, + "content": "#-block: ## this is treated literally!\n This is the other kind of block tag.\n It can have #italic{nested} block children.\n\n #-block: ## this is treated literally!\n #:\n\n #-block[prop]:label ## this is treated literally!\n Block starts/ends can have aribtrary labels attached.\n #:label ## this is treated literally!\n\n #-block[set='42']:label ## this is treated literally!\n Block starts/ends can have aribtrary labels attached.\n #:label ## this is treated literally!\n\n #-block[prop1, set: '42', prop2]:label ## this is treated literally!\n Block starts/ends can have aribtrary labels attached.\n #:label ## this is treated literally!\n#: ## this is treated literally!" + }, + "type": "tag", + "subtype": "literal" + }, + { + "value": [ + { + "value": "Some unicode escapes:", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "\"😸\" is a smiling cat face,", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "and \"\u000b\" is a vertical tab.", + "type": "text" + } + ], + "type": "paragraph" + } + ] +} diff --git a/data/golden/invalid-utf8/input.pro b/data/golden/invalid-utf8/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/invalid-utf8/input.pro @@ -0,0 +1,6 @@ +case: Invalid Utf-8 +--- + +The next paragraph contains the character #lit{0xC0}, which is not valid UTF-8. + +Invalid:À diff --git a/data/golden/invalid-utf8/output.json b/data/golden/invalid-utf8/output.json new file mode 100644 --- /dev/null +++ b/data/golden/invalid-utf8/output.json @@ -0,0 +1,52 @@ +{ + "content": [ + { + "value": [ + { + "value": "The next paragraph contains the character", + "type": "text" + }, + { + "type": "break" + }, + { + "value": { + "content": [ + { + "value": "0xC0", + "type": "text" + } + ], + "name": "lit", + "metadata": { + "settings": {}, + "properties": {} + } + }, + "subtype": "inline", + "type": "tag" + }, + { + "value": ", which is not valid UTF-8.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "Invalid:�", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "metadata": { + "settings": { + "case": "Invalid Utf-8" + }, + "properties": {} + } +} diff --git a/data/golden/line-break/input.pro b/data/golden/line-break/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/line-break/input.pro @@ -0,0 +1,7 @@ +case: Line breaks +--- +This is a paragraph ending in a Unix break. + +This is a paragraph ending in a Windows break. + +This is a paragraph ending in a classic Mac OS break. diff --git a/data/golden/line-break/output.json b/data/golden/line-break/output.json new file mode 100644 --- /dev/null +++ b/data/golden/line-break/output.json @@ -0,0 +1,37 @@ +{ + "content": [ + { + "value": [ + { + "value": "This is a paragraph ending in a Unix break.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This is a paragraph ending in a Windows break.", + "type": "text" + } + ], + "type": "paragraph" + }, + { + "value": [ + { + "value": "This is a paragraph ending in a classic Mac OS break.", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "metadata": { + "settings": { + "case": "Line breaks" + }, + "properties": {} + } +} diff --git a/data/golden/shebang/input.pro b/data/golden/shebang/input.pro new file mode 100644 --- /dev/null +++ b/data/golden/shebang/input.pro @@ -0,0 +1,5 @@ +#!psh +--- + +This document has a shebang at the start. +Make sure we can parse it. diff --git a/data/golden/shebang/output.json b/data/golden/shebang/output.json new file mode 100644 --- /dev/null +++ b/data/golden/shebang/output.json @@ -0,0 +1,24 @@ +{ + "content": [ + { + "value": [ + { + "value": "This document has a shebang at the start.", + "type": "text" + }, + { + "type": "break" + }, + { + "value": "Make sure we can parse it.", + "type": "text" + } + ], + "type": "paragraph" + } + ], + "metadata": { + "settings": {}, + "properties": {} + } +} diff --git a/prosidy.cabal b/prosidy.cabal new file mode 100644 --- /dev/null +++ b/prosidy.cabal @@ -0,0 +1,117 @@ +cabal-version: 2.4 +name: prosidy +version: 1.5.0.1 +synopsis: A simple language for writing documents. +license: MPL-2.0 +license-file: LICENSE +author: James Alexander Feldman-Crough +maintainer: alex@fldcr.com +copyright: ©2020 to James Alexander Feldman-Crough +category: Language +data-dir: data +data-files: golden/**/*.pro, golden/**/*.json +extra-source-files: CHANGELOG, README.pro + +tested-with: + GHC == 8.4.4 + , GHC == 8.6.5 + , GHC == 8.8.1 + +description: + Prosidy is a small language for writing documents. + More details are on its site at https://prosidy.org. + +source-repository head + type: git + location: https://git.fldcr.com/prosidy/prosidy + +flag fatal-warnings + description: Turns all warnings into errors. Used in CI. + default: False + manual: True + +------------------------------------------------------------------------------- +library + default-language: Haskell2010 + hs-source-dirs: src + + ghc-options: + -Wall + -Wno-name-shadowing + + if flag(fatal-warnings) + ghc-options: + -Werror + + exposed-modules: + Prosidy + , Prosidy.Optics + , Prosidy.Optics.Source + , Prosidy.Optics.Types + , Prosidy.Parse + , Prosidy.Source + , Prosidy.Types + , Prosidy.Types.Assoc + , Prosidy.Types.Key + , Prosidy.Types.Set + , Prosidy.Types.Series + + other-modules: + Prosidy.Optics.Internal + , Prosidy.Compat + + build-depends: + base >= 4.11 && < 5 + , aeson >= 1.4 && < 1.5 + , bytestring >= 0.10 && < 0.11 + , binary >= 0.8 && < 0.9 + , containers >= 0.6 && < 0.7 + , contravariant >= 1.5 && < 1.6 + , deepseq >= 1.4 && < 1.5 + , hashable >= 1.2 && < 1.4 + , megaparsec >= 7.0 && < 8.1 + , profunctors >= 5.3 && < 5.6 + , tagged >= 0.8 && < 0.9 + , text >= 1.2 && < 1.3 + , transformers >= 0.5 && < 0.6 + , vector >= 0.12 && < 0.13 + , unordered-containers >= 0.2 && < 0.3 + +------------------------------------------------------------------------------- +test-suite prosidy-test + default-language: Haskell2010 + type: exitcode-stdio-1.0 + hs-source-dirs: test + + ghc-options: + -Wall + -Wno-name-shadowing + + main-is: + Prosidy/Test.hs + + other-modules: + Prosidy.Test.Source + , Prosidy.Test.Types + , Prosidy.Test.Parse + , Paths_prosidy + + autogen-modules: + Paths_prosidy + + build-depends: + base + , prosidy + , aeson >= 1.4 && < 1.5 + , aeson-pretty >= 0.8 && < 0.9 + , aeson-diff >= 1.1 && < 1.2 + , bytestring >= 0.10 && < 0.11 + , containers >= 0.6 && < 0.7 + , directory >= 1.3 && < 1.4 + , filepath >= 1.4 && < 1.5 + , tasty >= 1.2 && < 1.3 + , tasty-ant-xml >= 1.1 && < 1.2 + , tasty-golden >= 2.3 && < 2.4 + , tasty-hunit >= 0.10 && < 1.11 + , tasty-quickcheck >= 0.10 && < 0.11 + , text >= 1.2 && < 1.3 diff --git a/src/Prosidy.hs b/src/Prosidy.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy.hs @@ -0,0 +1,19 @@ +{-| +Module : Prosidy +Description : Top-level module exporting types, parsing functions, and optics. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +module Prosidy (module X) where + +import Prosidy.Source as X + ( Source + , Location + , Line + , Column + , Offset + ) +import Prosidy.Optics as X +import Prosidy.Parse as X +import Prosidy.Types as X diff --git a/src/Prosidy/Compat.hs b/src/Prosidy/Compat.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Compat.hs @@ -0,0 +1,8 @@ +{-# LANGUAGE CPP #-} +module Prosidy.Compat + ( MonadFail(..) + ) where + +#if __GLASGOW_HASKELL__ < 808 +import Control.Monad.Fail (MonadFail(..)) +#endif diff --git a/src/Prosidy/Optics.hs b/src/Prosidy/Optics.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Optics.hs @@ -0,0 +1,11 @@ +{-| +Module : Prosidy.Optics +Description : A convenience module which reëxports optics definitions. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +module Prosidy.Optics (module X) where + +import Prosidy.Optics.Source as X +import Prosidy.Optics.Types as X diff --git a/src/Prosidy/Optics/Internal.hs b/src/Prosidy/Optics/Internal.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Optics/Internal.hs @@ -0,0 +1,116 @@ +{-| +Module : Prosidy.Optics.Internal +Description : Internal implementations of common Optics functions, removing a dependency on lens. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE RankNTypes #-} +{-# LANGUAGE TupleSections #-} +module Prosidy.Optics.Internal + ( module Prosidy.Optics.Internal + , Profunctor(..) + , Choice(..) + , Strong(..) + , Contravariant(..) + ) +where + +import Data.Profunctor ( Profunctor(..) + , Choice(..) + , Strong(..) + ) +import Data.Functor.Const ( Const(..) ) +import Data.Monoid ( First(..) + , Endo(..) + ) +import Data.Functor.Identity ( Identity(..) ) +import Data.Tagged ( Tagged(..) ) +import Data.Functor.Contravariant ( Contravariant(..) ) + +type Optic p f s t a b = p a (f b) -> p s (f t) +type Iso s t a b = forall p f . (Profunctor p, Functor f) => Optic p f s t a b +type Lens s t a b = forall p f . (Strong p, Functor f) => Optic p f s t a b +type Prism s t a b + = forall p f . (Choice p, Applicative f) => Optic p f s t a b +type Affine s t a b + = forall p f . (Choice p, Strong p, Applicative f) => Optic p f s t a b +type Traversal s t a b = forall f . (Applicative f) => Optic (->) f s t a b + +type Optic' p f s a = Optic p f s s a a +type Iso' s a = Iso s s a a +type Lens' s a = Lens s s a a +type Prism' s a = Prism s s a a +type Affine' s a = Affine s s a a +type Traversal' s a = Traversal s s a a + +type Getter s a = forall f . (Functor f, Contravariant f) => Optic' (->) f s a + +iso :: (s -> a) -> (b -> t) -> Iso s t a b +iso get set = dimap get (fmap set) +{-# INLINE iso #-} + +lens :: (s -> a) -> (s -> b -> t) -> Lens s t a b +lens get set = dimap into outof . second' + where + into x = (x, get x) + outof (x, f) = fmap (set x) f +{-# INLINE lens #-} + +prism :: (b -> t) -> (s -> Either t a) -> Prism s t a b +prism set get = dimap get rhs . right' where rhs = either pure (fmap set) +{-# INLINE prism #-} + +prism' :: (b -> s) -> (s -> Maybe a) -> Prism s s a b +prism' set get = dimap lhs rhs . right' + where + lhs x = maybe (Left x) Right (get x) + rhs = either pure (fmap set) +{-# INLINE prism' #-} + +affine :: (s -> Either t a) -> (s -> b -> t) -> Affine s t a b +affine get set = dimap lhs rhs . right' . second' + where + lhs x = fmap (x, ) $ get x + rhs = either pure (\(x, f) -> set x <$> f) +{-# INLINE affine #-} + +affine' :: (s -> Maybe a) -> (s -> b -> s) -> Affine s s a b +affine' get set = dimap lhs rhs . right' . second' + where + lhs x = maybe (Left x) (Right . (x, )) $ get x + rhs (Left x ) = pure x + rhs (Right (x, f)) = set x <$> f +{-# INLINE affine' #-} + +nullAffine :: Affine s s a b +nullAffine = affine' (const Nothing) const +{-# INLINE nullAffine #-} + +to :: (s -> a) -> Getter s a +to k = dimap k (contramap k) +{-# INLINE to #-} + +view :: Lens s t a b -> s -> a +view f = getConst . f Const +{-# INLINE view #-} + +views :: Traversal s t a b -> s -> [a] +views f = flip appEndo [] . getConst . f (Const . Endo . (:)) +{-# INLINE views #-} + +preview :: Optic (->) (Const (First a)) s t a b -> s -> Maybe a +preview f = getFirst . getConst . f (Const . First . Just) +{-# INLINE preview #-} + +over :: Optic (->) Identity s t a b -> (a -> b) -> s -> t +over t f = runIdentity . t (Identity . f) +{-# INLINE over #-} + +assign :: Optic' (->) Identity s a -> a -> s -> s +assign t = over t . const +{-# INLINE assign #-} + +review :: Prism' s a -> a -> s +review p = runIdentity . unTagged . p . Tagged . Identity +{-# INLINE review #-} diff --git a/src/Prosidy/Optics/Source.hs b/src/Prosidy/Optics/Source.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Optics/Source.hs @@ -0,0 +1,84 @@ +{-| +Module : Prosidy.Optics.Source +Description : Profunctor optics over Prosidy types. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +module Prosidy.Optics.Source + ( -- * Classy optics; implementable on all types with a location + HasLocation(..) + , offset + -- ** Read-only optics. + , column + , line + , source + + -- * Conversion utilities + , sparse + ) +where + +import Prosidy.Types +import Prosidy.Source +import Prosidy.Optics.Internal + +-- | A classy optic for selecting the 'Location' from a value. Note that +-- 'location' is affine: a 'Location' can't be attached to a value which does +-- not -- already have one, and not all values with an instance of +-- 'HasLocation' have a location. +class HasLocation t where + location :: Affine' t Location + +instance HasLocation Location where + location = id + {-# INLINE location #-} + +instance HasLocation (Tag a) where + location = affine' tagLocation (\d l -> d { tagLocation = Just l }) + {-# INLINE location #-} + +instance HasLocation (Region a) where + location = affine' regionLocation (\d l -> d { regionLocation = Just l }) + {-# INLINE location #-} + +instance HasLocation Paragraph where + location = + affine' paragraphLocation (\d l -> d { paragraphLocation = Just l }) + {-# INLINE location #-} + +-- | Focus on the 'Offset' from a value parsed from a source file. If the +-- 'Offset' is modified, note that the resulting 'column' and 'line' will /also/ be +-- modified as they are denormalizations of this value. +offset :: HasLocation l => Affine' l Offset +offset = location . sparse . lens + sparseLocationOffset + (\sl x -> sl { sparseLocationOffset = x }) +{-# INLINE offset #-} + +-- | Fetch the 'Column' from a value parsed from a source file. Modifications +-- are not allowed as the 'offset' and 'line' may become inconsistent. +column + :: (HasLocation l, Contravariant f, Applicative f) => Optic' (->) f l Column +column = location . to locationColumn +{-# INLINE column #-} + +-- | Fetch the 'Line' from a value parsed from a source file. Modifications +-- are not allowed as the 'offset' and 'column' may become inconsistent. +line :: (HasLocation l, Contravariant f, Applicative f) => Optic' (->) f l Line +line = location . to locationLine +{-# INLINE line #-} + +-- | Fetch the 'Source' a value was parsed from. Modifications are not allowed +-- as the 'line', 'offset', and 'column' may become inconsistent. +source + :: (HasLocation l, Contravariant f, Applicative f) => Optic' (->) f l Source +source = location . to locationSource +{-# INLINE source #-} + +-- | An isomorphism between 'Location' and 'SparseLocation'. This is allowed +-- because although a 'Location' has strictly more data than a 'SparseLocation', +-- those values are denormalizations of values within 'SparseLocation'. +sparse :: Iso' Location SparseLocation +sparse = iso stripLocation enrichLocation +{-# INLINE sparse #-} diff --git a/src/Prosidy/Optics/Types.hs b/src/Prosidy/Optics/Types.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Optics/Types.hs @@ -0,0 +1,233 @@ +{-| +Module : Prosidy.Optics.Types +Description : Optics for definitions in 'Prosidy.Types'. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE LambdaCase #-} +{-# LANGUAGE RankNTypes #-} +{-# LANGUAGE TypeFamilies #-} +module Prosidy.Optics.Types + ( -- * Classy optics + -- ** Items with 'Metadata' + HasMetadata(..) + , properties + , settings + , hasProperty + , atSetting + -- ** Items wrapping content + , HasContent(..) + -- * Accessors for fields not otherwise covered + , tag + -- * Conversion between 'Tag's and 'Region's. + , tagged + -- * Prisms on 'Block' contexts + , _BlockTag + , _BlockLiteral + , _BlockParagraph + -- * Prisms on 'Inline' contexts + , _InlineTag + , _Text + , _Break + -- * Optics on common types + , key + , _Assoc + , _NonEmpty + , _Series + , _SeriesNE + , _Set + ) +where + +import Prosidy.Types +import Prosidy.Types.Assoc ( toHashMap + , fromHashMap + ) +import Prosidy.Types.Series ( toSeq + , fromSeq + , toSeqNE + , fromSeqNE + ) +import Prosidy.Types.Set ( toHashSet + , fromHashSet + ) +import Prosidy.Optics.Internal + +import Data.Text ( Text ) +import Data.Sequence ( Seq ) +import Data.HashMap.Strict ( HashMap ) +import Data.HashSet ( HashSet ) + +import qualified Data.HashMap.Strict as HM +import qualified Data.HashSet as HS + +------------------------------------------------------------------------------- +-- | A classy optic for focusing on items with 'Metadata', including 'Tag's, +-- 'Region's, and 'Document's. +class HasMetadata t where + metadata :: Lens' t Metadata + +instance HasMetadata Document where + metadata = lens documentMetadata (\d m -> d { documentMetadata = m }) + {-# INLINE metadata #-} + +instance HasMetadata (Tag a) where + metadata = lens tagMetadata (\d m -> d { tagMetadata = m }) + {-# INLINE metadata #-} + +instance HasMetadata (Region a) where + metadata = lens regionMetadata (\d m -> d { regionMetadata = m }) + {-# INLINE metadata #-} + +instance HasMetadata Metadata where + metadata = id + {-# INLINE metadata #-} + +-- | Fetch all properties from items which contain metadata. +properties :: HasMetadata m => Lens' m (Set Key) +properties = + metadata . lens metadataProperties (\m p -> m { metadataProperties = p }) +{-# INLINABLE properties #-} +{-# SPECIALIZE INLINE properties :: Lens' Metadata (Set Key) #-} +{-# SPECIALIZE INLINE properties :: Lens' Document (Set Key) #-} +{-# SPECIALIZE INLINE properties :: Lens' (Tag a) (Set Key) #-} +{-# SPECIALIZE INLINE properties :: Lens' (Region a) (Set Key) #-} + +-- | Fetch all settings defined on items which contain metadata. +settings :: HasMetadata m => Lens' m (Assoc Key Text) +settings = + metadata . lens metadataSettings (\m s -> m { metadataSettings = s }) +{-# INLINABLE settings #-} +{-# SPECIALIZE INLINE settings :: Lens' Metadata (Assoc Key Text) #-} +{-# SPECIALIZE INLINE settings :: Lens' Document (Assoc Key Text) #-} +{-# SPECIALIZE INLINE settings :: Lens' (Tag a) (Assoc Key Text) #-} +{-# SPECIALIZE INLINE settings :: Lens' (Region a) (Assoc Key Text) #-} + +-- | Check if a property is attached to an item with metadata. Using this +-- optic as a setter will add a property if set to 'True' and remove the +-- property when set to 'False'. +hasProperty :: HasMetadata m => Key -> Lens' m Bool +hasProperty k = properties . _Set . lens + (HS.member k) + (\hs b -> (if b then HS.insert else HS.delete) k hs) +{-# INLINE hasProperty #-} + +-- | Select a setting from an item attached to metadata. Returns 'Nothing' if +-- no value is set. +atSetting :: HasMetadata m => Key -> Lens' m (Maybe Text) +atSetting k = settings . _Assoc . lens + (HM.lookup k) + (\hm x -> maybe (HM.delete k) (HM.insert k) x hm) +{-# INLINE atSetting #-} + +------------------------------------------------------------------------------- +-- | An optic for selecting children of an item in a recursive structure. +class HasContent t where + -- | The type of /all/ of the children collectively. For instance, + -- @type Content Document = Series Block@, as 'Document' has zero or more + -- contained 'Block's. + type Content t + content :: Lens' t (Content t) + +instance HasContent Document where + type Content Document = Series Block + content = lens documentContent (\d c -> d { documentContent = c }) + {-# INLINE content #-} + +instance HasContent (Tag a) where + type Content (Tag a) = a + content = lens tagContent (\t c -> t { tagContent = c }) + {-# INLINE content #-} + +instance HasContent (Region a) where + type Content (Region a) = a + content = lens regionContent (\t c -> t { regionContent = c }) + {-# INLINE content #-} + +instance HasContent Paragraph where + type Content Paragraph = SeriesNE Inline + content = lens paragraphContent (\t c -> t { paragraphContent = c }) + {-# INLINE content #-} + +------------------------------------------------------------------------------- +-- | Focus on the name of a 'Tag'. +tag :: Lens' (Tag a) Key +tag = lens tagName (\t n -> t { tagName = n }) +{-# INLINE tag #-} + +------------------------------------------------------------------------------- +-- | Focus on the inner 'Region' of 'Tag's with a name. This can be used to +-- filter 'Tag's to a specific subset for manipulation. +tagged :: Key -> Prism' (Tag a) (Region a) +tagged k = prism' (regionToTag k) + $ \tag -> if tagName tag == k then Just $ tagToRegion tag else Nothing +{-# INLINE tagged #-} + +------------------------------------------------------------------------------- +-- | Focus only on block tags. +_BlockTag :: Prism' Block BlockTag +_BlockTag = prism' BlockTag $ \case + BlockTag t -> Just t + _ -> Nothing + +-- | Focus only on paragraphs' +_BlockParagraph :: Prism' Block Paragraph +_BlockParagraph = prism' BlockParagraph $ \case + BlockParagraph p -> Just p + _ -> Nothing + +-- | Focus only on literal tags. +_BlockLiteral :: Prism' Block LiteralTag +_BlockLiteral = prism' BlockLiteral $ \case + BlockLiteral t -> Just t + _ -> Nothing + +-- | Focus only on inline tags. +_InlineTag :: Prism' Inline InlineTag +_InlineTag = prism' InlineTag $ \case + InlineTag t -> Just t + _ -> Nothing + +-- | Focus only on text nodes. +_Text :: Prism' Inline Text +_Text = prism' InlineText $ \case + InlineText t -> Just t + _ -> Nothing + +-- | Focus only on breaks. +_Break :: Prism' Inline () +_Break = prism' (const Break) $ \case + Break -> Just () + _ -> Nothing + +------------------------------------------------------------------------------- +-- | A Prism from 'Text' into a valid 'Key'. +key :: Prism' Text Key +key = prism' rawKey (either (const Nothing) Just . makeKey) +{-# INLINE key #-} + +-- | An isomorphism between Prosidy's 'Assoc' wrapper and 'HashMap'. +_Assoc :: Iso (Assoc k v) (Assoc k' v') (HashMap k v) (HashMap k' v') +_Assoc = iso toHashMap fromHashMap +{-# INLINE _Assoc #-} + +-- | A prism between possibly-empty and non-empty containers. +_NonEmpty :: Prism' (Series a) (SeriesNE a) +_NonEmpty = prism' (fromSeq . toSeqNE) (fromSeqNE . toSeq) +{-# INLINE _NonEmpty #-} + +-- | An isomorpism between Prosidy's 'Series' wrapper and 'Seq'. +_Series :: Iso (Series a) (Series b) (Seq a) (Seq b) +_Series = iso toSeq fromSeq +{-# INLINE _Series #-} + +-- | A prism from a non-empty 'Seq' into a 'SeriesNE'. +_SeriesNE :: Prism' (Seq a) (SeriesNE a) +_SeriesNE = prism' toSeqNE fromSeqNE +{-# INLINE _SeriesNE #-} + +-- | An isomorphism between Prosidy's 'Set' wrapper and 'HashSet'. +_Set :: Iso (Set a) (Set b) (HashSet a) (HashSet b) +_Set = iso toHashSet fromHashSet +{-# INLINE _Set #-} diff --git a/src/Prosidy/Parse.hs b/src/Prosidy/Parse.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Parse.hs @@ -0,0 +1,451 @@ +{- | +Module : Prosidy.Parse +Description : Parse raw text into Prosidy documents +Copyright : (c) James Alexander Feldman-Crough, 2019 +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE ApplicativeDo #-} +{-# LANGUAGE TypeApplications #-} +{-# LANGUAGE OverloadedStrings #-} + +module Prosidy.Parse + ( -- * Parsing Prosidy types from 'Data.Text.Text' + parseDocument + , parseDocumentMetadata + -- * Reading & parsing Prosidy files + , readDocument + , readDocumentMetadata + -- * Errors + , Failure(..) + , prettyFailure + ) +where + +import Prosidy.Compat +import Prelude hiding ( fail ) + +import Prosidy.Types +import Prosidy.Source +import Prosidy.Types.Key ( isValidKeyHead + , isValidKeyTail + , unsafeMakeKey + ) +import Prosidy.Types.Series ( fromSeqNE + , toSeqNE + , fromSeq + ) + +import Text.Megaparsec hiding ( token + , sourceName + ) +import Text.Megaparsec.Char ( char + , string + ) + +import qualified Data.Char as Char +import qualified Data.Text.Lazy as Text.Lazy +import qualified Data.HashMap.Strict as HashMap +import qualified Data.Sequence as Seq +import qualified Data.List as List +import qualified Data.HashSet as HashSet +import qualified Data.Text.Encoding as Text.Encoding +import qualified Text.Megaparsec.Char as Megaparsec +import qualified Data.ByteString as ByteString + +import Control.Applicative ( Alternative ) +import Data.Bifunctor ( first ) +import Text.Megaparsec.Char.Lexer ( hexadecimal ) +import Data.Functor ( ($>) ) +import Data.Foldable ( fold + , traverse_ + ) +import Control.Monad ( MonadPlus + , void + ) +import Data.Text ( Text ) +import Data.Void ( Void ) +import Control.Exception ( Exception + , throwIO + ) +import Control.Monad.Trans.Reader ( ReaderT(..) ) + +{- + If you run into errors, use the following combinator to get Megaparsec to + print out its state. + + > import qualified Text.Megaparsec.Debug + > + > dbg :: Show a => String -> P a -> P a + > dbg txt (P (ReaderT r)) = P . ReaderT $ \src -> + > Text.Megaparsec.Debug.dbg txt $ r src +-} + +------------------------------------------------------------------------------- +-- | Parses a Prosidy 'Document' from its source. +-- +-- The 'FilePath' parameter is only used for error reporting. +parseDocument :: FilePath -> Text -> Either Failure Document +parseDocument path = runP doc . makeSource path + +-- | Reads a Prosidy 'Document' from the given 'FilePath'. +-- +-- Errors will be thrown as exceptions. Use 'parseDocument' for a pure +-- implementation. +readDocument :: FilePath -> IO Document +readDocument filepath = do + bytes <- ByteString.readFile filepath + either throwIO pure . parseDocument filepath $ Text.Encoding.decodeUtf8With + (\_ _ -> Just '\65533') + bytes + +------------------------------------------------------------------------------- +-- | Parses a Prosidy document's header 'Metadata' from source, stopping when the +-- header ends. +-- +-- The 'FilePath' parameter is only used for error reporting. +parseDocumentMetadata :: FilePath -> Text -> Either Failure Metadata +parseDocumentMetadata path = runP docMetadata . makeSource path + +-- | Reads a Prosidy document's 'Metadata' header from the given 'FilePath'. +-- +-- Errors will be thrown as exceptions. Use 'parseDocumentMetadata' for a pure +-- implementation. +readDocumentMetadata :: FilePath -> IO Metadata +readDocumentMetadata filepath = do + bytes <- ByteString.readFile filepath + either throwIO pure + . parseDocumentMetadata filepath + $ Text.Encoding.decodeUtf8 bytes + +------------------------------------------------------------------------------- +-- | A parsing error. +-- +newtype Failure = Failure (ParseErrorBundle Text Void) + deriving newtype (Exception, Show) + +-- | Pretty-print a 'Failure' into a message acceptable for displaying to +-- users. +prettyFailure :: Failure -> String +prettyFailure (Failure e) = errorBundlePretty e + +------------------------------------------------------------------------------- +newtype P a = P (ReaderT Source (Parsec Void Text) a) + deriving newtype (Functor, Applicative, Alternative, Monad, MonadFail, MonadPlus, MonadParsec Void Text) + +type MetadataItem = (Key, Maybe Text) + +runP :: P a -> Source -> Either Failure a +runP (P (ReaderT r)) src = + first Failure $ parse (r src) (sourceName src) (sourceText src) + +------------------------------------------------------------------------------- +doc :: P Document +doc = do + header <- docMetadata + body <- Series . Seq.fromList <$> many block + eof + pure $ Document header body + +------------------------------------------------------------------------------- +docMetadata :: P Metadata +docMetadata = do + -- try to read a shebang as the _very_ first line of a document. + optional_ $ string "#!" *> skipManyTill anySingle (void newlineOrEOF) + -- skip past any blank lines at the start of the document + skipSpaces *> skipMany endOfLine + -- read each metadata item as a line-wide token + items <- many docMetadataItem + -- stop when we hit three dashes, alone, on a line + docMetadataEnd + pure $ foldMap itemToMetadata items + +docMetadataEnd :: P () +docMetadataEnd = do + void $ string "---" + try $ do + skipSpaces + newlineOrEOF + skipSpaces + skipMany endOfLine + +docMetadataItem :: P MetadataItem +docMetadataItem = do + itemKey <- key + itemVal <- optional $ do + metaItemSep + option "" text <* skipSpaces + endOfLines + pure (itemKey, itemVal) + +------------------------------------------------------------------------------- +block :: P Block +block = choice + [ BlockTag <$> blockTag + , BlockLiteral <$> literalTag + , BlockParagraph <$> paragraph + ] + +blockTag :: P BlockTag +blockTag = do + t <- genericTag (void $ string "#-") blockTagContents + emptyLines + pure t + +blockTagContents :: P (Series Block) +blockTagContents = choice [ifBraces, ifBlock, ifNothing] + where + ifBraces = annotateSource $ fmap + (foldMap $ \x src -> + Series . Seq.singleton . BlockParagraph $ Paragraph x src + ) + (token tagParagraph) + ifBlock = Series . Seq.fromList <$> withBlockDelimiters + (emptyLines *> many block) + ifNothing = skipSpaces *> endOfLine $> mempty + +literalTag :: P LiteralTag +literalTag = genericTag (void $ string "#=") $ do + close <- blockTagDelim (void $ optional_ comment *> newlineOrEOF) + literalBody close + +literalBody :: P () -> P Text +literalBody end = do + literalLines <- manyTill literalLine (try $ skipSpaces *> end) + emptyLines + pure $ Text.Lazy.toStrict . Text.Lazy.intercalate "\n" $ literalLines + +literalLine :: P Text.Lazy.Text +literalLine = do + line <- takeWhileP (Just "literal text") $ \ch -> ch /= '\r' && ch /= '\n' + newlineOrEOF + pure $ Text.Lazy.fromStrict line + +blockTagDelim :: P () -> P (P ()) +blockTagDelim slurp = do + char ':' + maybeLabel <- optional keyLike + skipSpaces <* slurp + pure $ do + string "#:" + traverse_ string maybeLabel + skipSpaces + +withBlockDelimiters :: P a -> P a +withBlockDelimiters parser = do + close <- blockTagDelim endOfLine + parser <* close + +------------------------------------------------------------------------------- +inline :: P Inline +inline = choice [InlineTag <$> inlineTag, InlineText <$> fragment] + +inlineTag :: P InlineTag +inlineTag = genericTag sigil . option mempty $ orEmpty tagParagraph + where + orEmpty = fmap $ maybe mempty (fromSeq . toSeqNE) + sigil = try $ do + void $ char '#' + void . lookAhead $ satisfy isValidKeyHead + +------------------------------------------------------------------------------- +paragraph :: P Paragraph +paragraph = annotateSource $ paragraphLike >>= maybe + (fail "empty paragraph encountered") + (pure . Paragraph) + +paragraphLike :: P (Maybe (SeriesNE Inline)) +paragraphLike = do + ppLines <- paragraphLine `sepEndBy1` endOfLine + emptyLines + pure . fromSeqNE . Seq.fromList $ List.intercalate [Break] ppLines + +paragraphLine :: P [Inline] +paragraphLine = do + headItem <- inline + tailItem <- many paragraphInline + skipSpaces + pure $ headItem : tailItem + +paragraphInline :: P Inline +paragraphInline = (paragraphSpacer $> Break) <|> inline + +paragraphSpacer :: P () +paragraphSpacer = try $ do + skipSpaces1 + notFollowedBy $ void (string "##") <|> newlineOrEOF + +tagParagraph :: P (Maybe (SeriesNE Inline)) +tagParagraph = between start end $ option Nothing paragraphLike + where + start = char '{' *> skipSpaces *> emptyLines + end = skipSpaces *> emptyLines *> char '}' + +------------------------------------------------------------------------------- +genericTag :: P () -> P a -> P (Tag a) +genericTag sigilParser bodyParser = annotateSource $ do + sigilParser + thisName <- unsafeMakeKey <$> keyLike + thisMetadata <- meta + thisContent <- bodyParser + pure $ Tag thisName thisMetadata thisContent + +meta :: P Metadata +meta = + option mempty + $ between start end + $ foldMap itemToMetadata + <$> metaItem + `sepEndBy` metaSep + where + start = do + char '[' + skipSpaces + skipMany endOfLine + end = char ']' + +metaItem :: P MetadataItem +metaItem = do + itemKey <- key <* emptyLines + itemVal <- optional $ do + metaItemSep <* emptyLines + option "" quotedText + skipSpaces <* emptyLines + pure (itemKey, itemVal) + +metaSep :: P () +metaSep = do + void $ char ',' + skipSpaces + emptyLines + +------------------------------------------------------------------------------- +escape :: P Char +escape = label "escape sequence" $ do + void $ char '\\' + choice + [ oneOf @[] "#{}[]:='\"\\" + , char 'n' $> '\n' + , char 't' $> '\t' + , char 'r' $> '\r' + , char 'u' *> unicodeEscape + ] + +unicodeEscape :: P Char +unicodeEscape = Char.chr <$> hexadecimal + +keyLike :: P Text +keyLike = do + void . lookAhead $ satisfy isValidKeyHead + takeWhile1P (Just "key") isValidKeyTail + +key :: P Key +key = token $ unsafeMakeKey <$> keyLike + +metaItemSep :: P () +metaItemSep = token . void $ char ':' <|> char '=' + +quotedText :: P Text +quotedText = do + delim <- char '\'' <|> char '"' + parts <- many $ choice + [ Text.Lazy.singleton <$> escape + , Text.Lazy.fromStrict <$> takeWhile1P + (Just "quoted text") + (\ch -> ch /= delim && ch /= '\\') + ] + void $ char delim + skipSpaces + pure . Text.Lazy.toStrict . fold $ parts + +fragment :: P Text +fragment = text + +text :: P Text +text = do + parts <- word `sepBy1` textSpace + pure . Text.Lazy.toStrict . Text.Lazy.intercalate " " $ parts + +textSpace :: P () +textSpace = try $ do + skipSpaces1 + notFollowedBy $ void (char '#') <|> newlineOrEOF + +word :: P Text.Lazy.Text +word = fmap fold . some $ choice + [ Text.Lazy.singleton <$> escape + , Text.Lazy.fromStrict <$> takeWhile1P + (Just "plain text") + (\ch -> not $ HashSet.member ch reserved || Char.isSpace ch) + ] + where reserved = HashSet.fromList "#{}\\" + +------------------------------------------------------------------------------- +comment :: P () +comment = label "comment" $ do + void $ string "##" + void $ skipManyTill anySingle (lookAhead newlineOrEOF) + +endOfLine :: P () +endOfLine = + -- This rule is a bit hairy! Specifically, there was a bug at the end of + -- a file that ended with a comment and no trailing newline. + -- + -- Because we use `endOfLine` in repeat productions (many, some), + -- endOfLine _has_ to consume input to prevent looping forever. + -- In order to satisfy this: + -- + -- 1. If its the end of a file, then we _must_ consume a comment. + -- 2. If it's not the end of a file, then we _must_ consume at least + -- one newline. + commentThenNewline <|> commentThenEOF + where + commentThenEOF = comment <* eof + commentThenNewline = try $ do + optional_ comment + void Megaparsec.newline + skipSpaces + +endOfLines :: P () +endOfLines = skipSome endOfLine + +emptyLines :: P () +emptyLines = skipMany endOfLine + +spaceChar :: P () +spaceChar = do + notFollowedBy newlineOrEOF + void Megaparsec.spaceChar + +skipSpaces :: P () +skipSpaces = skipMany spaceChar + +skipSpaces1 :: P () +skipSpaces1 = skipSome spaceChar + +token :: P a -> P a +token = (<* skipSpaces) + +newlineOrEOF :: P () +newlineOrEOF = void Megaparsec.newline <|> eof + +------------------------------------------------------------------------------- +optional_ :: P a -> P () +optional_ = option () . void + +------------------------------------------------------------------------------- +itemToMetadata :: MetadataItem -> Metadata +itemToMetadata (k, Just v ) = Metadata mempty (Assoc $ HashMap.singleton k v) +itemToMetadata (k, Nothing) = Metadata (Set $ HashSet.singleton k) mempty + +annotateSource :: P (Maybe Location -> a) -> P a +annotateSource (P (ReaderT r)) = P . ReaderT $ \src -> do + offset <- Offset . fromIntegral <$> getOffset + result <- r src + sourceLoc <- maybe (fail sourceLocationError) pure $ getLocation offset src + pure . result $ Just sourceLoc + +sourceLocationError :: String +sourceLocationError = "UNEXPECTED: Failed to create a source location." diff --git a/src/Prosidy/Source.hs b/src/Prosidy/Source.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Source.hs @@ -0,0 +1,268 @@ +{-| +Module : Prosidy.Source +Description : Utilities for tracking source locaitons. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE DeriveAnyClass #-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE MultiWayIf #-} +{-# LANGUAGE MultiParamTypeClasses #-} +{-# LANGUAGE StrictData #-} +{-# LANGUAGE TupleSections #-} +{-# LANGUAGE TypeFamilies #-} +module Prosidy.Source + ( Source(..) + , Location + , SparseLocation(..) + , LineMap + , Offset(..) + , Line(..) + , Column(..) + , locationSource + , locationColumn + , locationLine + , locationOffset + , makeSource + , getSourceLine + , getLocation + , enrichLocation + , stripLocation + , lineOffsets + , lineToOffset + , offsetToLine + ) +where + +import Data.Hashable ( Hashable(..) ) +import Data.Vector.Unboxed ( Vector + , MVector + , Unbox + ) +import Data.Text ( Text ) +import GHC.Generics ( Generic ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary(..) ) +import Data.Aeson ( ToJSON(..) + , FromJSON(..) + ) +import Control.Monad ( guard ) + +import qualified Data.Text as T +import qualified Data.Vector.Unboxed as V +import qualified Data.Vector.Generic as VG +import qualified Data.Vector.Generic.Mutable as VGM + +-- | Information about Prosidy source file. +-- +-- The 'Show' instance for ths class does not include the 'LineMap' or 'Text' +-- fields, as those are rather noisy. +data Source = Source + { sourceName :: String + -- ^ The reported file-name of the 'Source'. + -- + -- When read from file handles, a non-filepath description such as + -- @"\"@ is typically chosen. + -- This field doesn't have semantic meaning, and should only be used to + -- enrich the output displayed to users. + , sourceText :: Text + -- ^ The full source, as 'Text'. + , sourceLineMap :: LineMap + -- ^ A mapping of the start position of each line in the 'Source'. + } + deriving stock (Eq, Generic) + deriving anyclass (Hashable, NFData, Binary) + +instance Show Source where + show (Source fp _ _) = "Source " <> show fp + +-- | Create a 'Source' from a descriptive name and a body. The source name is +-- typically a 'FilePath', but this is not guarenteed. For instance, when read +-- from standard-input, Prosidy chooses to name the source @\@. +makeSource :: String -> Text -> Source +makeSource name body = Source name body lineMap + where + lineMap = case T.foldl' lineMapFold (1, '\0', []) $ body of + (_, _, acc) -> LineMap . V.fromList . reverse $ acc + lineMapFold (ix, prev, acc) ch + | ch == '\n' && prev == '\r' = (succ ix, ch, Offset ix : drop 1 acc) + | ch == '\n' || ch == '\r' = (succ ix, ch, Offset ix : acc) + | otherwise = (succ ix, ch, acc) + +-- | Convert an 'Offset' into a 'Location'. +getLocation :: Offset -> Source -> Maybe Location +getLocation offset src = do + guard $ T.length (sourceText src) > fromEnum offset + Just . enrichLocation $ SparseLocation src offset + +-- | Fetch a single line from a source. +getSourceLine :: Line -> Source -> Maybe Text +getSourceLine line source = do + start <- fromEnum <$> lineToOffset line lineMap + let end = fromEnum <$> lineToOffset (succ line) lineMap + Just . maybe id (T.take . (flip (-) start)) end . T.drop start $ sourceText + source + where lineMap = sourceLineMap source + +-- | A location in a 'Source'. The line and column numbers of this type are not +-- attached to this type; convert to a 'Location' to access those values. +data SparseLocation = SparseLocation + { sparseLocationSource :: Source + -- ^ The 'Source' this location references. + , sparseLocationOffset :: Offset + -- ^ The position in the 'Source', counted by Unicode codepoints. + } + deriving stock (Show, Generic, Eq) + deriving anyclass (NFData, Binary, Hashable) + +-- | A location in a 'Source', with the line and column number computed lazily. +data Location = Location + { locationSource :: Source + -- ^ The 'Source' this location references. + , locationOffset :: Offset + -- ^ The position in the 'Source', counted by Unicode codepoints. + , locationLine :: ~Line + -- ^ The line number in the 'Source'. + , locationColumn :: ~Column + -- ^ The column number in the 'Source'. + } + deriving stock (Show, Generic, Eq) + deriving anyclass (NFData, Binary, Hashable) + +-- | Add lazily computed line and column number information to a +-- 'SparseLocation'. +enrichLocation :: SparseLocation -> Location +enrichLocation sl = Location { locationSource = source + , locationOffset = offset + , locationLine = line + , locationColumn = column + } + where + source = sparseLocationSource sl + lineMap = sourceLineMap source + offset@(~(Offset offsetN)) = sparseLocationOffset sl + line = offsetToLine offset lineMap + column = case lineToOffset line lineMap of + Just (Offset n) -> Column (offsetN - n) + Nothing -> Column 0 + +-- | Remove line and column number information from a 'Location'. +stripLocation :: Location -> SparseLocation +stripLocation l = SparseLocation { sparseLocationSource = locationSource l + , sparseLocationOffset = locationOffset l + } + +-- | A dense vector containing offsets poiting to the start of each line. That +-- is, the starting position of the third line of a file can be found at +-- position 2. +newtype LineMap = LineMap (Vector Offset) + deriving stock (Eq, Generic) + deriving newtype (Show, NFData) + +instance Binary LineMap where + get = fmap (LineMap . V.fromList) get + put (LineMap v) = put (V.toList v) + +instance Hashable LineMap where + hashWithSalt salt (LineMap v) = V.foldl' hashWithSalt salt v + +-- | Convert a 'LineMap' into a list of 'Offset's, corresponding to the first +-- character of a line. Note that the initial offset is omitted-- the offset at +-- index 0 will be the offset of the /second/ line. +lineOffsets :: LineMap -> [Offset] +lineOffsets (LineMap v) = V.toList v + +-- | Fetch the 'Offset' for the given 'Line'. Evaluates to 'Nothing' if the +-- given 'Line' does not appear in the LineMap +lineToOffset :: Line -> LineMap -> Maybe Offset +lineToOffset (Line 0 ) _ = Just $ Offset 0 +lineToOffset (Line nth) (LineMap xs) = xs V.!? fromIntegral (pred nth) + +-- | Fetch the 'Line' number for a given 'Offset'. Newlines will be attributed +-- the line that they terminate, rather than the line started immediately +-- afterwards. +offsetToLine :: Offset -> LineMap -> Line +offsetToLine offset (LineMap xs) = Line . fromIntegral $ go Nothing + 0 + (V.length xs) + where + go result min max + | min >= max + = maybe 0 succ result + | otherwise + = let nthIndex = ((max - min) `div` 2) + min + nthOffset = xs V.! nthIndex + in case nthOffset `compare` offset of + EQ -> succ nthIndex + LT -> go (Just nthIndex) (nthIndex + 1) max + GT -> go result min nthIndex + +-- | A line number. +-- +-- The 'Show' instance for 'Line' counts from one, while the internal +-- implementation counts from zero. +newtype Line = Line Word + deriving stock (Eq, Ord, Generic, Show) + deriving newtype (ToJSON, FromJSON, Enum) + deriving anyclass (Hashable, NFData, Binary) + +-- | A column number. +newtype Column = Column Word + deriving stock (Eq, Ord, Generic, Show) + deriving newtype (ToJSON, FromJSON, Enum) + deriving anyclass (Hashable, NFData, Binary) + +-- | An offset into a 'Source', counted by UTF-8 codepoint. +newtype Offset = Offset Word + deriving stock (Eq, Show, Ord, Generic) + deriving newtype (ToJSON, FromJSON, Enum) + deriving anyclass (Hashable, NFData, Binary) + +newtype instance MVector s Offset = MV_Offset (MVector s Word) + +instance VGM.MVector MVector Offset where + basicLength (MV_Offset m) = VGM.basicLength m + {-# INLINE basicLength #-} + + basicUnsafeSlice ix len (MV_Offset m) = + MV_Offset $ VGM.basicUnsafeSlice ix len m + {-# INLINE basicUnsafeSlice #-} + + basicOverlaps (MV_Offset x) (MV_Offset y) = VGM.basicOverlaps x y + {-# INLINE basicOverlaps #-} + + basicUnsafeNew len = MV_Offset <$> VGM.basicUnsafeNew len + {-# INLINE basicUnsafeNew #-} + + basicInitialize (MV_Offset v) = VGM.basicInitialize v + {-# INLINE basicInitialize #-} + + basicUnsafeRead (MV_Offset v) = fmap Offset <$> VGM.basicUnsafeRead v + {-# INLINE basicUnsafeRead #-} + + basicUnsafeWrite (MV_Offset v) ix (Offset w) = VGM.basicUnsafeWrite v ix w + {-# INLINE basicUnsafeWrite #-} + +newtype instance Vector Offset = V_Offset (Vector Word) + +instance VG.Vector Vector Offset where + basicUnsafeFreeze (MV_Offset v) = V_Offset <$> VG.basicUnsafeFreeze v + {-# INLINE basicUnsafeFreeze #-} + + basicUnsafeThaw (V_Offset v) = MV_Offset <$> VG.basicUnsafeThaw v + {-# INLINE basicUnsafeThaw #-} + + basicLength (V_Offset v) = VG.basicLength v + {-# INLINE basicLength #-} + + basicUnsafeSlice ix len (V_Offset v) = + V_Offset $ VG.basicUnsafeSlice ix len v + {-# INLINE basicUnsafeSlice #-} + + basicUnsafeIndexM (V_Offset v) ix = Offset <$> VG.basicUnsafeIndexM v ix + {-# INLINE basicUnsafeIndexM #-} + +instance Unbox Offset where diff --git a/src/Prosidy/Types.hs b/src/Prosidy/Types.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Types.hs @@ -0,0 +1,311 @@ +{-| +Module : Prosidy.Types +Description : A convenience module which reëxports type definitions & helpers. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE DeriveAnyClass #-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE DeriveFunctor #-} +{-# LANGUAGE DeriveFoldable #-} +{-# LANGUAGE DeriveTraversable #-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE FlexibleContexts #-} +{-# LANGUAGE FlexibleInstances #-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE OverloadedStrings #-} +{-# LANGUAGE StrictData #-} +{-# LANGUAGE TypeApplications #-} +{-# LANGUAGE UndecidableInstances #-} +module Prosidy.Types + ( -- * Documents + Document(..) + , documentToRegion + , regionToDocument + -- * Tags + , Tag(..) + , BlockTag + , InlineTag + , LiteralTag + , tagToRegion + , regionToTag + -- * Contextual enumerations + , Block(..) + , Inline(..) + -- * Paragraphs + , Paragraph(..) + -- * Common structures + , Metadata(..) + , Region(..) + -- * Utility wrappers + , module X + ) +where + +import Prosidy.Types.Assoc as X + ( Assoc(..) ) +import Prosidy.Types.Key as X + ( Key + , KeyError(..) + , InvalidCharacter + , makeKey + , rawKey + ) +import Prosidy.Types.Series as X + ( Series(..) + , SeriesNE + ) +import Prosidy.Types.Set as X + ( Set(..) ) +import Prosidy.Source ( Location ) + +import Data.Text ( Text ) +import GHC.Generics ( Generic ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary ) +import Data.Hashable ( Hashable ) +import Data.Aeson ( ToJSON(..) + , FromJSON(..) + , withObject + , (.:) + , (.=) + , object + , pairs + ) + +import qualified Data.Aeson as Aeson + +------------------------------------------------------------------------------- +-- | A sum type enumerating allowed types inside of a block context. +data Block = + BlockLiteral LiteralTag + | BlockParagraph Paragraph + | BlockTag BlockTag + deriving stock (Eq, Show, Generic) + deriving anyclass (Hashable, Binary, NFData) + +instance FromJSON Block where + parseJSON = withObject "block" $ \o -> do + ty <- o .: "type" + case ty :: Text of + "tag" -> do + subtype <- o .: "subtype" + case subtype :: Text of + "block" -> BlockTag <$> o .: "value" + "literal" -> BlockLiteral <$> o .: "value" + _ -> fail $ "unknown tag subtype: " <> show subtype + "paragraph" -> BlockParagraph <$> o .: "value" + _ -> fail $ "unknown block type: " <> show ty + +instance ToJSON Block where + toEncoding b = pairs . mconcat $ case b of + BlockLiteral t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("literal" :: Text) + , "value" .= t + ] + BlockParagraph p -> ["type" .= ("paragraph" :: Text), "value" .= p] + BlockTag t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("block" :: Text) + , "value" .= t + ] + + toJSON b = object $ case b of + BlockLiteral t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("literal" :: Text) + , "value" .= t + ] + BlockParagraph p -> ["type" .= ("paragraph" :: Text), "value" .= p] + BlockTag t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("block" :: Text) + , "value" .= t + ] + +------------------------------------------------------------------------------- +-- | A full Prosidy document. +data Document = Document + { documentMetadata :: Metadata + , documentContent :: Series Block + } + deriving stock (Eq, Show, Generic) + deriving anyclass (Hashable, NFData, Binary) + +instance FromJSON Document where + parseJSON = withObject "Document" + $ \o -> Document <$> o .: "metadata" <*> o .: "content" + +instance ToJSON Document where + toEncoding (Document md ct) = + pairs $ mconcat ["metadata" .= md, "content" .= ct] + + toJSON (Document md ct) = object ["metadata" .= md, "content" .= ct] + +-- | Convert a 'Document' to a 'Region'. The resulting 'Region' will never have +-- a 'Location' attached. +documentToRegion :: Document -> Region (Series Block) +documentToRegion (Document md ct) = Region md ct Nothing + +-- | Convert a 'Region' to a 'Document'. Any 'Location' on the 'Region' will be +-- discarded. +regionToDocument :: Region (Series Block) -> Document +regionToDocument (Region md ct _) = Document md ct + +------------------------------------------------------------------------------- +-- | A sum type enumerating allowed types inside of an inline context. +data Inline = + Break + -- ^ Spacing recorded between lines or on either side of an 'Inline' 'Tag'. + -- Although we could represent this as 'Text', Prosidy defines a special + -- node for this case so that authors in CJK languages (or other languages + -- without explicit spaces between words) may simply ignore these spaces + -- in their output. + | InlineTag InlineTag + -- ^ A 'Tag' which contains only 'Inline' items. These tags begin with the + -- @#@ sigil in source. + | InlineText Text + -- ^ A fragment of plain text. + deriving stock (Eq, Show, Generic) + deriving anyclass (Hashable, Binary, NFData) + +instance FromJSON Inline where + parseJSON = withObject "Inline" $ \o -> do + ty <- o .: "type" + case ty :: Text of + "break" -> pure Break + "tag" -> InlineTag <$> o .: "value" + "text" -> InlineText <$> o .: "value" + _ -> fail $ "unknown inline type: " <> show ty + +instance ToJSON Inline where + toEncoding i = pairs . mconcat $ case i of + Break -> ["type" .= ("break" :: Text), "value" .= Aeson.Null] + InlineTag t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("inline" :: Text) + , "value" .= t + ] + InlineText t -> ["type" .= ("text" :: Text), "value" .= t] + + toJSON i = object $ case i of + Break -> ["type" .= ("break" :: Text)] + InlineTag t -> + [ "type" .= ("tag" :: Text) + , "subtype" .= ("inline" :: Text) + , "value" .= t + ] + InlineText t -> ["type" .= ("text" :: Text), "value" .= t] + +------------------------------------------------------------------------------- +-- | A set of properties and settings, associated with a 'Region'. +-- +-- The namespaces of properties and settings are distinct; a property can share +-- a name with a setting without conflict. +data Metadata = Metadata + { metadataProperties :: Set Key + -- ^ Properties are a set of 'Key's with no associated value. + , metadataSettings :: Assoc Key Text + -- ^ Settings are 'Key's with an attached value. + } + deriving stock (Eq, Show, Generic) + deriving anyclass (NFData, Binary, Hashable) + +instance Monoid Metadata where + mempty = Metadata mempty mempty + +instance Semigroup Metadata where + Metadata p1 s1 <> Metadata p2 s2 = Metadata (p1 <> p2) (s1 <> s2) + +instance FromJSON Metadata where + parseJSON = withObject "Metadata" + $ \o -> Metadata <$> o .: "properties" <*> o .: "settings" + +instance ToJSON Metadata where + toEncoding (Metadata ps ss) = + pairs $ mconcat ["properties" .= ps, "settings" .= ss] + + toJSON (Metadata ps ss) = object ["properties" .= ps, "settings" .= ss] + +------------------------------------------------------------------------------- +-- | A non-empty collection of 'Inline' items. A 'Paragraph' represents the +-- border between block and inline contexts. All ancestors of a paragraph are +-- block items or a document, and all children are inline items. +data Paragraph = Paragraph + { paragraphContent :: SeriesNE Inline + , paragraphLocation :: Maybe Location + } + deriving stock (Eq, Show, Generic) + deriving anyclass (Hashable, NFData, Binary) + +instance FromJSON Paragraph where + parseJSON = fmap (flip Paragraph Nothing) . parseJSON + +instance ToJSON Paragraph where + toEncoding (Paragraph s _) = toEncoding s + toJSON (Paragraph s _) = toJSON s + +------------------------------------------------------------------------------- +-- | An untagged structural grouping of items with type @a@. Regions do not +-- occur in parsing. +data Region a = Region + { regionMetadata :: Metadata + , regionContent :: a + , regionLocation :: Maybe Location + } + deriving stock (Eq, Foldable, Functor, Show, Traversable, Generic) + deriving anyclass (Hashable, NFData, Binary) + +instance ToJSON a => ToJSON (Region a) where + toJSON (Region md ct _) = Aeson.object ["metadata" .= md, "content" .= ct] + +------------------------------------------------------------------------------- +-- | A 'Region', annotated with a tag name. +data Tag a = Tag + { tagName :: Key + , tagMetadata :: Metadata + , tagContent :: a + , tagLocation :: Maybe Location + } + deriving stock (Eq, Foldable, Functor, Show, Traversable, Generic) + deriving anyclass (Hashable, NFData, Binary) + +instance FromJSON a => FromJSON (Tag a) where + parseJSON = withObject "Tag" $ \o -> + Tag + <$> o + .: "name" + <*> o + .: "metadata" + <*> o + .: "content" + <*> pure Nothing + +instance ToJSON a => ToJSON (Tag a) where + toEncoding (Tag nm md ct _) = + pairs $ mconcat ["name" .= nm, "metadata" .= md, "content" .= ct] + + toJSON (Tag nm md ct _) = + object ["name" .= nm, "metadata" .= md, "content" .= ct] + +-- | A 'Tag' containing zero or more 'Block' items. +-- Specified in Prosidy source with the @#-@ sigil. +type BlockTag = Tag (Series Block) + +-- | A 'Tag' containing zero or more 'Inline' items. +-- Specified in Prosidy source with the @#@ sigil. +type InlineTag = Tag (Series Inline) + +-- | A 'Tag' containing a single plain-text item. +-- Specified in Prosidy source with the @#=@ sigil. +type LiteralTag = Tag Text + +-- | Convert a 'Tag' to a 'Region' by discarding the tag's name. +tagToRegion :: Tag a -> Region a +tagToRegion (Tag _ md ct loc) = Region md ct loc + +-- | Convert a 'Region' to a 'Tag' by providing a tag name. +regionToTag :: Key -> Region a -> Tag a +regionToTag name (Region md ct loc) = Tag name md ct loc diff --git a/src/Prosidy/Types/Assoc.hs b/src/Prosidy/Types/Assoc.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Types/Assoc.hs @@ -0,0 +1,54 @@ +{-| +Module : Prosidy.Types.Assoc +Description : An associative mapping of keys to values. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +module Prosidy.Types.Assoc (Assoc(..), asHashMap, fromHashMap, toHashMap) where + +import Data.HashMap.Strict ( HashMap ) +import GHC.Generics ( Generic ) +import Data.Aeson ( ToJSON(..) + , FromJSON(..) + ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary(..) ) +import Data.Hashable ( Hashable(..) ) + +import qualified Data.HashMap.Strict as HM + +-- | An associative mapping of keys to values. +-- +-- Currently implemented as a 'HashMap', this newtype wrapper allows us to: +-- +-- 1) Add non-orphan instances to the underlying structure. +-- 2) Change the underlying type if needed. +newtype Assoc k v = Assoc (HashMap k v) + deriving stock (Generic) + deriving newtype (Eq, Foldable, Functor, Show, ToJSON, FromJSON, NFData, Semigroup, Monoid, Hashable) + +instance (Eq k, Hashable k, Binary k, Binary v) => Binary (Assoc k v) where + get = Assoc . HM.fromList <$> get + + put (Assoc hm) = put $ HM.toList hm + +-- | Given a function which operates on a 'HashMap', return a function which +-- performs an equivalent transfromation on an 'Assoc'. +asHashMap + :: Functor f + => (HashMap k v -> f (HashMap k' v')) + -> Assoc k v + -> f (Assoc k' v') +asHashMap f (Assoc a) = Assoc <$> f a + +-- | Convert a 'HashMap' to an 'Assoc'. +fromHashMap :: HashMap k v -> Assoc k v +fromHashMap = Assoc + +-- | Convert an 'Assoc' to a 'HashMap'. +toHashMap :: Assoc k v -> HashMap k v +toHashMap (Assoc hm) = hm diff --git a/src/Prosidy/Types/Key.hs b/src/Prosidy/Types/Key.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Types/Key.hs @@ -0,0 +1,155 @@ +{-| +Module : Prosidy.Types.Key +Description : Definitions and helpers for 'Key'. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE DeriveAnyClass #-} +{-# LANGUAGE DeriveFoldable #-} +{-# LANGUAGE DeriveFunctor #-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE DeriveTraversable #-} +{-# LANGUAGE GADTSyntax #-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE LambdaCase #-} +{-# LANGUAGE OverloadedStrings #-} +{-# LANGUAGE PatternSynonyms #-} +{-# LANGUAGE RankNTypes #-} +{-# LANGUAGE StandaloneDeriving #-} +{-# LANGUAGE StrictData #-} +{-# LANGUAGE TemplateHaskell #-} +{-# LANGUAGE TypeFamilies #-} +{-# LANGUAGE ViewPatterns #-} +{-# LANGUAGE QuasiQuotes #-} +module Prosidy.Types.Key + ( -- * The 'Key' type. + Key + -- * Creating 'Key's and unwrapping them + , makeKey + , rawKey + , unsafeMakeKey + -- * Checking validity of raw text. + , isValidKeyHead + , isValidKeyTail + -- * Errors + , KeyError(..) + , InvalidCharacter(..) + ) +where + +import Data.Text ( Text ) +import Data.Aeson ( ToJSON(..) + , ToJSONKey(..) + , FromJSON(..) + , FromJSONKey(..) + ) +import GHC.Generics ( Generic ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary ) +import Data.Hashable ( Hashable ) +import Data.String ( IsString(..) ) +import Data.Foldable ( for_ ) +import Control.Monad ( unless ) +import Control.Exception ( Exception(..) + , throw + ) + +import qualified Data.Aeson as Aeson +import qualified Data.Char as Char +import qualified Data.Set as Set +import qualified Data.Text as Text + +-- | A 'Key' is an identifier used in tags, properties, and setting names. +newtype Key = Key Text + deriving stock (Generic) + deriving newtype (Binary, Eq, Hashable, NFData, Ord, Show, ToJSON, ToJSONKey) + +-- | 'Key' exposes an 'IsString' instance, but beware! Invalid strings will +-- throw a pure exception. +instance IsString Key where + fromString = either throw id . makeKey . Text.pack + +instance FromJSON Key where + parseJSON json = do + text <- parseJSON json + either (fail . displayException) pure $ makeKey text + +instance FromJSONKey Key where + fromJSONKey = + Aeson.FromJSONKeyTextParser + $ either (fail . displayException) pure + . makeKey + +-- | Create a new 'Key', checking its validity. +makeKey :: Text -> Either KeyError Key +makeKey rawText = case Text.unpack rawText of + [] -> Left EmptyKeyError + keyHead : keyTail + | isValidKeyHead keyHead -> do + for_ (zip [1 ..] keyTail) $ \(ix, ch) -> + unless (isValidKeyTail ch) + $ Left + . InvalidCharacterError + $ InvalidCharacter rawText ix ch + Right $ Key rawText + | otherwise -> Left . InvalidCharacterError $ InvalidCharacter + rawText + 0 + keyHead + +-- | Create a new 'Key' /without/ performing any checks. +unsafeMakeKey :: Text -> Key +unsafeMakeKey = Key +{-# INLINE unsafeMakeKey #-} + +-- | Convert a 'Key' into its 'Text' representation. +rawKey :: Key -> Text +rawKey (Key key) = key + +-- | Check if a character is suitable for use as the first character in a +-- 'Key'. +isValidKeyHead :: Char -> Bool +isValidKeyHead = (||) <$> Char.isAlphaNum <*> (== '_') + +-- | Check if a character is suitable for use as any character except the +-- first character in a 'Key'. +isValidKeyTail :: Char -> Bool +isValidKeyTail = not . invalid + where + invalid = (||) <$> Char.isSpace <*> (`Set.member` reserved) + reserved = Set.fromList "\\#{}[]:=," + +-- | Errors returned when creating invalid keys. +data KeyError = + InvalidCharacterError InvalidCharacter + -- ^ A character provided as a 'Key'\'s name was invalid. + | EmptyKeyError + -- ^ A string of length 0 was provided as a 'Key'\'s name. + deriving (Show, Eq) + +-- | Details for errors thrown when creating 'Key's with one or more invalid +-- characters. +data InvalidCharacter = InvalidCharacter + { -- | The full string provided as the 'Key's name. + invalidCharacterText :: Text + -- | The position of the invalid character. + , invalidCharacterPosition :: Word + -- | The exact character that was invalid. + , invalidCharacterCharacter :: Char + } + deriving (Show, Eq) + +instance Exception KeyError where + displayException EmptyKeyError = + "Cannot create a Key with a length of zero." + displayException (InvalidCharacterError (InvalidCharacter text nth ch)) = + unwords + [ "Cannot create a Key named " <> show text <> ":" + , "the character" + , show ch + , "at index" + , show nth + , "is not allowed." + ] diff --git a/src/Prosidy/Types/Series.hs b/src/Prosidy/Types/Series.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Types/Series.hs @@ -0,0 +1,109 @@ +{-| +Module : Prosidy.Types.Series +Description : A type of items occuring in a sequence. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +module Prosidy.Types.Series + ( -- * Possibly empty collections + Series(..) + , asSeq + , fromSeq + , toSeq + -- * Known non-empty collections + , SeriesNE + , fromSeqNE + , toSeqNE + ) +where + +import Data.Sequence ( Seq ) +import GHC.Generics ( Generic ) +import Data.Aeson ( ToJSON(..) + , FromJSON(..) + ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary(..) ) +import Data.Hashable ( Hashable(..) ) +import Data.Foldable ( toList + , foldl' + ) +import Control.Monad ( guard ) + +import qualified Data.Sequence as Seq + +-- | A newtype wrapper around a sequential collection. +-- +-- Currently, 'Series' is implemented as a 'Seq', but this is not guarenteed to +-- be true. +newtype Series a = Series (Seq a) + deriving stock (Generic, Show) + deriving newtype (Eq, Foldable, Functor, Applicative, ToJSON, FromJSON, NFData, Semigroup, Monoid) + +instance Binary a => Binary (Series a) where + get = Series . Seq.fromList <$> get + {-# INLINE get #-} + + put (Series xs) = put $ toList xs + {-# INLINE put #-} + +instance Hashable a => Hashable (Series a) where + hashWithSalt salt (Series xs) = foldl' hashWithSalt salt xs + +instance Traversable Series where + traverse f (Series xs) = Series <$> traverse f xs + +-- | A non-empty 'Series'. +newtype SeriesNE a = SeriesNE (Seq a) + deriving stock (Generic, Show) + deriving newtype (Eq, Foldable, Functor, Applicative, ToJSON, NFData, Semigroup, Monoid) + +instance Binary a => Binary (SeriesNE a) where + get = + maybe (error "SeriesNE must be non-empty") id + . fromSeqNE + . Seq.fromList + <$> get + {-# INLINE get #-} + + put (SeriesNE xs) = put $ toList xs + {-# INLINE put #-} + +instance FromJSON a => FromJSON (SeriesNE a) where + parseJSON value = do + inner <- parseJSON value + guard (not $ null inner) + pure $ SeriesNE inner + +instance Hashable a => Hashable (SeriesNE a) where + hashWithSalt salt (SeriesNE xs) = foldl' hashWithSalt salt xs + +instance Traversable SeriesNE where + traverse f (SeriesNE xs) = SeriesNE <$> traverse f xs + +-- | Given a function which operates on a 'Seq', return a function which +-- operates on a 'Series'. +asSeq :: Functor f => (Seq a -> f (Seq b)) -> Series a -> f (Series b) +asSeq f (Series s) = Series <$> f s + +-- | Convert a 'Seq' into a 'Series'. +fromSeq :: Seq a -> Series a +fromSeq = Series + +-- | Convert a 'Series' into a 'Seq'. +toSeq :: Series a -> Seq a +toSeq (Series s) = s + +-- | Convert a non-empty 'Seq' into a 'SeriesNE'. +fromSeqNE :: Seq a -> Maybe (SeriesNE a) +fromSeqNE s | null s = Nothing +fromSeqNE s | otherwise = Just (SeriesNE s) + +-- | Convert a 'SeriesNE' into a 'Seq'. The returned 'Seq' is guarenteed to +-- always contain at least one element. +toSeqNE :: SeriesNE a -> Seq a +toSeqNE (SeriesNE a) = a diff --git a/src/Prosidy/Types/Set.hs b/src/Prosidy/Types/Set.hs new file mode 100644 --- /dev/null +++ b/src/Prosidy/Types/Set.hs @@ -0,0 +1,59 @@ +{-| +Module : Prosidy.Types.Set +Description : An unordered collection of unique items. +Copyright : ©2020 James Alexander Feldman-Crough +License : MPL-2.0 +Maintainer : alex@fldcr.com +-} +{-# LANGUAGE GeneralizedNewtypeDeriving #-} +{-# LANGUAGE DerivingStrategies #-} +{-# LANGUAGE DeriveGeneric #-} +{-# LANGUAGE DeriveAnyClass #-} +module Prosidy.Types.Set (Set(..), asHashSet, fromHashSet, toHashSet) where + +import Data.HashSet ( HashSet ) +import GHC.Generics ( Generic ) +import Data.Aeson ( FromJSONKey + , ToJSONKey + , ToJSON(..) + , FromJSON(..) + ) +import Control.DeepSeq ( NFData ) +import Data.Binary ( Binary(..) ) +import Data.Hashable ( Hashable(..) ) +import qualified Data.HashSet as HS +import qualified Data.HashMap.Strict as HM + +-- | A newtype wrapper around an unordered collection of unique elements. +-- +-- Currently, this is implemented as a wrapper around a 'HashSet'. +newtype Set a = Set (HashSet a) + deriving stock (Generic) + deriving newtype (Eq, Foldable, Show, NFData, Semigroup, Monoid, Hashable) + +instance (Hashable a, Eq a, ToJSONKey a) => ToJSON (Set a) where + toJSON (Set hs) = toJSON $ foldMap (flip HM.singleton True) hs + toEncoding (Set hs) = toEncoding $ foldMap (flip HM.singleton True) hs + +instance (Hashable a, Eq a, FromJSONKey a) => FromJSON (Set a) where + parseJSON json = do + m <- parseJSON json + pure . Set . HM.keysSet $ HM.filter id m + +instance (Eq a, Hashable a, Binary a) => Binary (Set a) where + get = Set . HS.fromList <$> get + + put (Set s) = put $ HS.toList s + +-- | Given a function which operates on 'HashSet's, return a function which +-- performs the same operation on a 'Set'. +asHashSet :: Functor f => (HashSet a -> f (HashSet b)) -> Set a -> f (Set b) +asHashSet f (Set s) = Set <$> f s + +-- | Convert a 'Set' to a 'HashSet'. +toHashSet :: Set a -> HashSet a +toHashSet (Set s) = s + +-- | Convert a 'HashSet' to a 'Set'. +fromHashSet :: HashSet a -> Set a +fromHashSet = Set diff --git a/test/Prosidy/Test.hs b/test/Prosidy/Test.hs new file mode 100644 --- /dev/null +++ b/test/Prosidy/Test.hs @@ -0,0 +1,21 @@ +import Test.Tasty +import Test.Tasty.Runners.AntXML (antXMLRunner) +import qualified Prosidy.Test.Source +import qualified Prosidy.Test.Types +import qualified Prosidy.Test.Parse + +main :: IO () +main = tests >>= defaultMainWithIngredients ingredients + where + ingredients = defaultIngredients ++ + [ antXMLRunner + ] + +tests :: IO TestTree +tests = do + parseTests <- Prosidy.Test.Parse.tests + pure $ testGroup "test" + [ Prosidy.Test.Source.tests + , Prosidy.Test.Types.tests + , parseTests + ] diff --git a/test/Prosidy/Test/Parse.hs b/test/Prosidy/Test/Parse.hs new file mode 100644 --- /dev/null +++ b/test/Prosidy/Test/Parse.hs @@ -0,0 +1,64 @@ +{-# LANGUAGE OverloadedStrings #-} +{-# LANGUAGE ScopedTypeVariables #-} +module Prosidy.Test.Parse (tests) where + +import qualified Prosidy + +import Test.Tasty +import Test.Tasty.Golden.Advanced (goldenTest) +import qualified Data.ByteString as BS +import qualified Data.ByteString.Lazy as BS.Lazy +import qualified System.Directory as Dir +import System.FilePath (()) +import Control.Exception (handle, displayException) + +import qualified Paths_prosidy as Paths +import qualified Data.Aeson as Aeson +import qualified Data.Aeson.Diff as Aeson.Diff +import qualified Data.Aeson.Encode.Pretty as Aeson.Pretty + +import qualified Data.Text.Lazy as Text.Lazy +import Data.Text.Lazy.Encoding (decodeUtf8With) + +tests :: IO TestTree +tests = do + goldenDir <- ( "golden") <$> Paths.getDataDir + goldenTests <- Dir.listDirectory goldenDir + pure . testGroup "parse" $ makeTest goldenDir <$> goldenTests + +makeTest :: FilePath -> String -> TestTree +makeTest goldenDir name = + goldenTest name + (catchErrors getJSON) + (catchErrors parsePro) + compareDocs + writeGolden + where + getJSON :: IO Aeson.Value + getJSON = do + bytes <- BS.readFile $ goldenDir name "output.json" + case Aeson.eitherDecode' . BS.Lazy.fromStrict $ bytes of + Left e -> fail $ "Failed to parse JSON: " <> e + Right ok -> pure ok + + parsePro :: IO Aeson.Value + parsePro = Aeson.toJSON <$> + Prosidy.readDocument (goldenDir name "input.pro") + + compareDocs :: Aeson.Value -> Aeson.Value -> IO (Maybe String) + compareDocs gold test | gold == test = pure Nothing + compareDocs gold test = + let + diff = Aeson.Diff.diff (Aeson.toJSON gold) (Aeson.toJSON test) + pretty = Aeson.Pretty.encodePretty diff + in + pure . Just . Text.Lazy.unpack $ + "Golden test failed. The diff is included below:\n" <> + decodeUtf8With (\_ _ -> Just '\65533') pretty + + writeGolden :: Aeson.Value -> IO () + writeGolden doc = BS.Lazy.writeFile (goldenDir name "output.json") $ + Aeson.Pretty.encodePretty doc + +catchErrors :: IO a -> IO a +catchErrors = handle $ \(e :: Prosidy.Failure) -> fail $ displayException e diff --git a/test/Prosidy/Test/Source.hs b/test/Prosidy/Test/Source.hs new file mode 100644 --- /dev/null +++ b/test/Prosidy/Test/Source.hs @@ -0,0 +1,120 @@ +{-# LANGUAGE OverloadedStrings #-} +module Prosidy.Test.Source (tests) where + +import Test.Tasty +import Test.Tasty.HUnit +import Test.Tasty.QuickCheck +import Data.Foldable (toList, for_) +import Data.IORef +import System.IO.Unsafe +import Control.Exception (evaluate) + +import qualified Prosidy.Source as PS +import qualified Data.Text as T + +tests :: TestTree +tests = testGroup "source" + [ testEmptyLineMap + , testSimpleLineMap + , testLineDelimiters + , testLocation + , testLocationLazy + , propLineOffset + ] + +testEmptyLineMap :: TestTree +testEmptyLineMap = testCase "empty" $ do + let source = PS.makeSource "" "" + assertBool "An empty source generates an empty line map." $ + null (PS.lineOffsets $ PS.sourceLineMap source) + +testSimpleLineMap :: TestTree +testSimpleLineMap = testCase "simple" $ do + let sourceLines = + [ "This source file contains a few lines." + , "Some, line the next, are empty." + , "" + , "Multiple consecutive empty lines are fine as well." + , "" + , "" + , "終わり" + ] + source = PS.makeSource "" $ T.unlines sourceLines + toList (PS.lineOffsets $ PS.sourceLineMap source) @?= + fmap PS.Offset [39, 71, 72, 123, 124, 125, 129] + for_ (zip [PS.Line 0 ..] sourceLines) $ \(lineNumber, line) -> + PS.getSourceLine lineNumber source @?= Just (line <> "\n") + +testLineDelimiters :: TestTree +testLineDelimiters = testCase "endings" $ do + let source = PS.makeSource "" "abc\ndef\rghi\r\njkl" + lines = PS.sourceLineMap source + PS.getSourceLine (PS.Line 0) source @?= Just "abc\n" + PS.lineToOffset (PS.Line 0) lines @?= Just (PS.Offset 0) + PS.getSourceLine (PS.Line 1) source @?= Just "def\r" + PS.lineToOffset (PS.Line 1) lines @?= Just (PS.Offset 4) + PS.getSourceLine (PS.Line 2) source @?= Just "ghi\r\n" + PS.lineToOffset (PS.Line 2) lines @?= Just (PS.Offset 8) + PS.getSourceLine (PS.Line 3) source @?= Just "jkl" + PS.lineToOffset (PS.Line 3) lines @?= Just (PS.Offset 13) + PS.getSourceLine (PS.Line 4) source @?= Nothing + +testLocation :: TestTree +testLocation = testCase "location" $ do + let source = PS.makeSource "" "ab\ncd\nef" + checkLocation offset line column = do + let loc = PS.getLocation (PS.Offset offset) source + fmap PS.locationOffset loc @?= Just (PS.Offset offset) + fmap PS.locationLine loc @?= Just (PS.Line line) + fmap PS.locationColumn loc @?= Just (PS.Column column) + checkLocation 0 0 0 + checkLocation 1 0 1 + checkLocation 2 0 2 + checkLocation 3 1 0 + checkLocation 4 1 1 + checkLocation 5 1 2 + checkLocation 6 2 0 + checkLocation 7 2 1 + PS.getLocation (PS.Offset 8) source @?= Nothing + +testLocationLazy :: TestTree +testLocationLazy = testCase "lazy" $ do + let source = PS.makeSource "" "abc\ndef" + Just loc = PS.getLocation (PS.Offset 5) source + (line, checkLine) <- checkEvaluated (PS.locationLine loc) + (col, checkCol) <- checkEvaluated (PS.locationColumn loc) + assertBool "line is unevaluated" . not =<< checkLine + assertBool "col is unevaluated" . not =<< checkCol + _ <- evaluate line + assertBool "line is evaluated" =<< checkLine + assertBool "col is unevaluated" . not =<< checkCol + _ <- evaluate col + assertBool "line is evaluated" =<< checkLine + assertBool "col is evaluated" =<< checkCol + +propLineOffset :: TestTree +propLineOffset = testProperty "line-and-offset" $ + forAll gen $ \(source, initialOffset) -> + let lineMap = PS.sourceLineMap source + line = PS.offsetToLine initialOffset lineMap + Just offset = PS.lineToOffset line lineMap + line' = PS.offsetToLine offset lineMap + in (initialOffset >= offset) .&&. (line === line') + where + gen = do + text <- T.pack <$> genChar + initialOffset <- elements [PS.Offset 0 .. toEnum (if T.null text then 0 else T.length text - 1)] + pure (PS.makeSource "" text, initialOffset) + genChar = listOf $ frequency + [ (10, elements ['a' .. 'z'] ) + , (4, pure ' ') + , (1, pure '\n') + ] + +{-# NOINLINE checkEvaluated #-} +checkEvaluated :: a -> IO (a, IO Bool) +checkEvaluated val = do + ref <- newIORef False + pure ( unsafePerformIO (writeIORef ref True) `seq` val + , readIORef ref + ) diff --git a/test/Prosidy/Test/Types.hs b/test/Prosidy/Test/Types.hs new file mode 100644 --- /dev/null +++ b/test/Prosidy/Test/Types.hs @@ -0,0 +1,78 @@ +{-# LANGUAGE OverloadedStrings #-} +module Prosidy.Test.Types (tests) where + +import Test.Tasty +import Test.Tasty.QuickCheck + +import qualified Prosidy.Types.Key as Key +import qualified Prosidy.Types.Series as Series +import qualified Data.Text as Text +import qualified Data.Sequence as Seq + +tests :: TestTree +tests = testGroup "types" + [ testKey + , testSeriesNE + ] + +testKey :: TestTree +testKey = testGroup "key" + [ testProperty "valid" $ forAll validGen $ \raw -> + fmap Key.rawKey (Key.makeKey raw) === Right raw + , testProperty "invalid-head" $ forAll invalidHeadGen $ \raw -> + case Key.makeKey raw of + Right _ -> label "key creation should fail" False + Left Key.EmptyKeyError -> label "key should be nonempty" $ property False + Left (Key.InvalidCharacterError (Key.InvalidCharacter raw' nth ch)) -> + conjoin + [ label "error string matches input" $ raw === raw' + , label "invalid at index 0" $ nth === 0 + , label "char is invalid" $ not (Key.isValidKeyHead ch) + ] + , testProperty "invalid-tail" $ forAll invalidTailGen $ \raw -> + case Key.makeKey raw of + Right _ -> label "key creation should fail" False + Left Key.EmptyKeyError -> label "key should be nonempty" $ property False + Left (Key.InvalidCharacterError (Key.InvalidCharacter raw' nth ch)) -> + conjoin + [ label "error string matches input" $ raw === raw' + , label "invalid at index > 0" $ nth =/= 0 + , label "char is invalid" $ not (Key.isValidKeyTail ch) + ] + , testProperty "empty" $ + case Key.makeKey "" of + Right _ -> label "key creation should fail" False + Left Key.EmptyKeyError -> property () + Left (Key.InvalidCharacterError _) -> label "input should be empty" False + ] + where + arbHead = arbitraryUnicodeChar `suchThat` Key.isValidKeyHead + arbTail = listOf $ arbitraryUnicodeChar `suchThat` Key.isValidKeyTail + + validGen = do + keyHead <- arbHead + keyTail <- arbTail + pure . Text.pack $ keyHead : keyTail + + invalidHeadGen = do + keyHead <- arbitraryUnicodeChar `suchThat` (not . Key.isValidKeyHead) + keyTail <- arbTail + pure . Text.pack $ keyHead : keyTail + + invalidTailGen = do + keyHead <- arbHead + keyTail0 <- arbTail + keyTail1 <- elements "\n\t\v\\#{}[]:=, " + keyTail2 <- arbTail + pure . Text.pack $ keyHead : keyTail0 ++ keyTail1 : keyTail2 + +testSeriesNE :: TestTree +testSeriesNE = testGroup "SeriesNE" + [ testProperty "nonempty" $ + forAll genSeq1 $ \xs -> + fmap Series.toSeqNE (Series.fromSeqNE xs) === Just xs + , testProperty "empty" $ + Series.fromSeqNE (mempty :: Seq.Seq Word) === Nothing + ] + where + genSeq1 = fmap Seq.fromList . listOf1 $ (arbitrary :: Gen Word)