Yalafi

--------------------------
- yalafi: fixed issue 23 (error recovery for missing \] or \}),
added test to tests/test\_error.py
- yalafi.shell
- yalafi/shell/proofreader.py: fixed issue 14
- added output mode 'xml-b' for XML report, addresses issue 16
- added some fields to JSON output
- added option --lt-command, issue 19
- added options --enable, --disablecategories, --enablecategories
(issue 21)
- added ALE linter component lty.vim
- README.md
- new section on related projects, added reference to vscode-ltex
- added application of Vim plug-in vim-LanguageTool
- added application of Vim plug-in ALE
- shifted screenshots to directory figs/
- this file: added dates to versions

--------------------------
- yalafi/shell/proofreader.py
- fixed issue 12
- more verbose error messages, if start of LT fails
- replaced screenshots for Vim and Emacs application
- updated README.md

--------------------------
- now hopefully close to 1.0.0 ;-)
- yalafi.shell
- closed issue 7, option --as-server: emulate an LT server with
integrated LaTeX filter
- added interface for Emacs-langtool
- added output mode --output json
- added option --no-config
- closed issue 9 (a hack): better highlighting, if a single backslash
is tagged that actually starts a macro name
(as before for HTML output only)
- removed option --t2t-lang (just use 'en' on unknown language)
- subdivided yalafi/shell/shell.py into several files
- README.md: updated

--------------------------
- yalafi.shell
- option --lt-options: added passing of -l / --language to HTML request
- fixed problem with Java under Cygwin: always first go to LT's
directory, then start Java
- fixed Issue 6
- closed Issue 2, except: cannot gracefully recover from wrong parameters
of Python definitions in Macro() etc. in yalafi/paramters.py
- fixed Issue 3, added Bash script yalafi-grammarous
- closed Issue 4
- fixed Issue 5
- yalafi/mathparser.py: slightly improved character position tracking
in displayed equations
- speed optimisation
- for large input files, modifications at the start of the token list are
expensive, producing roughly quadratic run-time of the complete filter
- we now operate on reversed lists in class scanner.Buffer and in method
parser.Parser.remove\_pure\_action\_lines()
- result: faster, and run-time increases quasi linearly with file size
- renamed 'opts' parameter of Macro() etc. to 'defaults'
- README.md: updated

--------------------------
- yalafi.shell
- added output in XML format for vim-grammarous
- removed option --html
- added option '--output mode', mode value in plain, html, xml
- changed option name --plain to --plain-input
- added option --lt-directory
- README.md: updated

--------------------------
- added replication of punctuation marks at specified \\item labels
- verbatim environment now can be removed or replaced like other environments
with appropriate entry in Parser.environment\_defs
- added default labels for environment enumerate
- yalafi/tex2txt.py: option --ienc is effective for file from --defs, too
- README.md: updated
<br><br>
- yalafi/scanner.py: tested use of regular expression; only a bit faster for
short texts, but much slower for large texts