aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2021-03-20Move yamlMetaBlock from Markdown reader to T.P.Readers.Metadata.John MacFarlane2-22/+22
2021-03-20Markdown reader: export `yamlMetaBlock`.John MacFarlane1-17/+23
[API change] This will allow us to parse YAML metadata blocks in other readers, potentially.
2021-03-20Text.Pandoc.Parsing: remove F type synonym.John MacFarlane5-9/+9
Muse and Org were defining their own F anyway, with their own state. We therefore move this definition to the Markdown reader.
2021-03-20T.P.Readers.Metadata: made `yamlBsToMeta`, `yamlBsToRefs` polymorphic...John MacFarlane1-15/+15
on the parser state, instead of requiring ParserState. [API change]
2021-03-20RST writer: use NonEmpty for init, last.John MacFarlane1-8/+12
2021-03-20Include Header.Attr.attributes as XML attributes on sectionErik Rask2-2/+82
Add key-value pairs found in the attributes list of Header.Attr as XML attributes on the corresponding section element. Any key name not allowed as an XML attribute name is dropped, as are keys with invalid values where they are defined as enums in DocBook, and xml:id (for DocBook 5)/id (for DocBook 4) to not intervene with computed identifiers.
2021-03-20Add date on changelogJohn MacFarlane1-1/+1
2021-03-20Update man page and manual date.John MacFarlane2-3/+3
2021-03-20Bump to 2.13, update changelogJohn MacFarlane2-6/+48
2021-03-20T.P.Shared: remove `backslashEscapes`, `escapeStringUsing`.John MacFarlane8-47/+77
[API change] These are inefficient association list lookups. Replace with more efficient functions in the writers that used them (with 10-25% performance improvements in haddock, org, rtf, texinfo writers).
2021-03-19Tests: Use getExecutablePath from base...John MacFarlane4-5/+3
avoiding the need to depend on the executable-path package.
2021-03-19Fix fallback to default partials on templates.John MacFarlane1-0/+4
If the directory containing a template does not contain the partial, it should be sought in the default data files. Closes #7164.
2021-03-19Tests: factor out setupEnvironment in Test.Helpers.John MacFarlane3-25/+26
This avoids code duplication between Command and Old.
2021-03-19Fix finding of data files from test programs.John MacFarlane2-2/+5
Apparently Cabal sets a `pandoc_datadir` environment variable so that the data files will be sought in the source directory rather than in the final destination (where they aren't yet installed). So we no longer need to set `--data-dir` in the tests. We just need to make sure `pandoc_datadir` is set in the environment when we call the program in the test suite. This will fix the issue with loading of pandoc.lua when pandoc is built with `-embed_data_files`, reported in #7163. Closes #7163.
2021-03-19cabal.project: move flags: to top level (not under pandoc:).John MacFarlane1-5/+1
If it's under the pandoc stanza, it can't be overriden on the command line, it seems. Also we remove ghc-options.
2021-03-19Hlint suggestion.John MacFarlane1-2/+3
2021-03-19CI: test with different flags.John MacFarlane1-3/+9
2021-03-19Narrow version bounds for skylighting, citeproc, and texmath.John MacFarlane1-4/+4
This reduces the chance that tests will fail due to behavior changes in one of these dependencies. (See e.g. #7163)
2021-03-19T.P.Shared: Remove ToString, ToText typeclasses [API change].John MacFarlane2-24/+4
T.P.Parsing: revise type of readWithM so that it takes a Text rather than a polymorphic ToText value. These typeclasses were there to ease the transition from String to Text. They are no longer needed, and they may clash with more useful versions under the same name. This will require a bump to 2.13.
2021-03-19Use skylighting 0.10.5.John MacFarlane2-4/+4
This fixes a bad regression in Haskell syntax highlighting.
2021-03-19Protect partial uses of maximum with NonEmpty.John MacFarlane21-86/+108
2021-03-19Use NonEmpty instead of minimumDef.John MacFarlane4-8/+9
2021-03-19Docx reader: Don't reimplement NonEmpty.John MacFarlane1-5/+1
2021-03-18Use minimumDef instead of minimum (partial function).John MacFarlane4-9/+8
2021-03-18Require safe >= 0.3.18 and remove cpp.John MacFarlane2-6/+1
2021-03-18Rewrite a foldl1 as a foldl'.John MacFarlane1-1/+5
2021-03-18Remove another foldr1 partial function use.John MacFarlane1-5/+6
2021-03-18T.P.Readers.Odt.StyleReader: rewrite foldr1 use as foldr.John MacFarlane1-5/+6
This avoids a partial function.
2021-03-18Don't bake in extra stack size to the executables.John MacFarlane1-1/+1
I don't think this is necessary; stack overflows generally indicate a code problem and should be fixed (and have been when reported).
2021-03-18Minor revisions to provisional changelog.John MacFarlane1-6/+4
2021-03-18Bump to 2.12.1 and update changelog.John MacFarlane2-1/+94
2021-03-18Use -A8m for default rtsopts for benchmarkJohn MacFarlane1-1/+1
2021-03-18Revert the -j related changes in make_artifacts.sh.John MacFarlane1-3/+3
This led to very slow build times.
2021-03-18Change -A256m to -A8m in GHCOPTSJohn MacFarlane1-1/+1
2021-03-18pandoc.cabal: bake in -A8m to rtsopts.John MacFarlane1-1/+1
This reserves a larger allocation area and reduces GC, speeding up execution.
2021-03-18build-arm.sh : remove strict checking on first ssh accessJohn MacFarlane1-0/+2
2021-03-18make_artifacts.sh: try using -j for cabal rather than -j4 ghc-options.John MacFarlane2-5/+5
2021-03-17Use random 1.2.0 in stack.yaml.John MacFarlane1-0/+3
2021-03-17Docx writer: make nsid in abstractNum deterministic.John MacFarlane34-6/+4
Previously we assigned a random number (though in a deterministic way). But changes in the random package mean we get different results now on different architectures, even with the same random seed. We don't need random values; so now we just assign a value based on the list number id, which is guaranteed to be unique to the list marker.
2021-03-17Update bounds for random (#7156)Alexey Kuleshevich1-1/+1
2021-03-17Fix benchmark in ci.John MacFarlane1-1/+1
2021-03-17Benchmark improvements.John MacFarlane4-18/+13
* Build `+RTS -A256m -RTS` into default ghc-options for benchmark, so we don't have to specify this separately on the command line. This is necessary to get accurate benchmark results; otherwise we are largely measuring garbage collecting, some not related to the current benchmark. * Switch back from gauge to tasty-bench. * Allow specifying BASELINE file in 'make bench' for comparison (otherwise the latest is chosen by default). * Remove obsolete reference to weigh-pandoc from CONTRIBUTING.md. * Remove `-Rghc-timing` from 'make bench'.
2021-03-17Makefile: Add comment documenting benchmark settings.John MacFarlane1-0/+3
2021-03-17Fix regression with `tex_math_backslash` in Markdown reader.John MacFarlane1-1/+1
Added regression test. Closes #7155.
2021-03-17Add test for #7155.John MacFarlane1-0/+15
2021-03-17Remove unnecessary tasty-bench from extra-depsJohn MacFarlane1-1/+0
2021-03-17Revert "Makefile: add --static to quick target"John MacFarlane1-1/+1
This reverts commit 6075b17ae11e5f1670d5d6e715431c4242dfd943.
2021-03-16Makefile: add --static to quick targetJohn MacFarlane1-1/+1
2021-03-16Disable redundant brackets check in .hlint.yaml.John MacFarlane1-8/+2
2021-03-16Increase heap space in runtime for benchmarks.John MacFarlane4-4/+8
Otherwise we're essentially benchmarking garbage collecting, which can give very inconsistent results.
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-16 10:36:56 +0000