* lisp/org-element.el (org-element-normalize-contents): Fix return
value when any line after the first has no indentation.
* testing/lisp/test-org-element.el (test-org-element/normalize-contents):
Add test.
* lisp/org-element.el (org-element--current-element): Check that
planning info and property drawers are correctly located before
parsing them.
* testing/lisp/test-org-element.el (test-org-element/planning-parser):
(test-org-element/property-drawer-parser): Add tests.
* lisp/org-element.el (org-element--object-lex): Do not re-match past
1-character long radio links.
* testing/lisp/test-org-element.el (test-org-element/link-parser): Add
test.
Reported-by: Daniel Clemente <n142857@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/109861>
* lisp/org-element.el (org-element-class): New function.
(org-element-map):
(org-element-interpret-data):
* lisp/org-footnote.el (org-footnote--allow-reference-p):
* lisp/org-src.el (org-src--on-datum-p):
* lisp/ox-odt.el (org-odt-footnote-reference):
(org-odt-table-cell):
* lisp/ox.el (org-export-data):
(org-export-expand): Use new function.
* testing/lisp/test-org-element.el (test-org-element/class): New test.
Using generic `org-element-class' allows to handle unknown, i.e.,
pseudo, object or element types. It also reduces code duplication in
`org-element-interpret-data' and `org-export-data', preventing, e.g.,
bugs as the one fixed in c58e1b5.
* lisp/org-element.el (org-element-link-interpreter): % in description
are confused with format string placeholders. Escape them so as to
fix the error.
* testing/lisp/test-org-element.el (test-org-element/link-interpreter):
Add test.
Reported-by: Daniel Clemente <n142857@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/109878>
* lisp/org-element.el (org-element-link-parser): Add :format property.
Tiny refactoring.
* testing/lisp/test-org-element.el (test-org-element/link-interpreter):
Add one test. Update some others.
Since the link format is lost during parsing, the interpreter can hardly
handle nested links. Indeed, in that case, the inner link is
interpreted as a bracket link, thus breaking the outer link.
Another option could be to guess a safe format for the link to be
interpreted. E.g.,
- any inner link could default to angle format unless it contains ">",
it which case it would become a plain link;
- other links would have the bracket format, unless they contain "][" or
"]]".
In any case, defining :format is less error prone and is similar to what
entities and sub/super-scripts have for curly brackets.
Reported-by: Thibault Marin <thibault.marin@gmx.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/109623>
* lisp/org-element.el (org-element--parse-paired-brackets): New function.
(org-element-inline-babel-call-parser):
(org-element-inline-src-block-parser): Use new function.
* testing/lisp/test-org-element.el (test-org-element/inline-src-block-parser):
Add test.
Reported-by: Charles Millar <millarc@verizon.net>
<http://permalink.gmane.org/gmane.emacs.orgmode/107629>
* lisp/org-element.el (org-element-context): Make sure point is on tag
line before searching for it.
* testing/lisp/test-org-element.el (test-org-element/context): Add test.
Small refactoring.
Reported-by: Kaushal Modi <kaushal.modi@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/105018>
* lisp/org-element.el (org-element-src-block-parser):
(org-element-example-block-parser): Do not remove any indentation when
parsing the element.
* testing/lisp/test-org-element.el (test-org-element/src-block-parser):
* testing/lisp/test-org-element.el (test-org-element/src-block-parser):
Update test.
Properly handling indentation requires knowing about
`org-src-preserve-indentation''s value. This is out of the scope of the
parser.
* lisp/org-element.el (org-element-context): Do not look for objects
within TODO keyword, priority cookie, comment keyword or tags.
* lisp/org.el (org-activate-tags): Fix regexp.
* testing/lisp/test-org-element.el (test-org-element/context): Add test.
Reported-by: John Kitchin <jkitchin@andrew.cmu.edu>
* lisp/org-element.el (org-element--set-regexps):
(org-element-footnote-definition-interpreter):
(org-element-footnote-reference-parser):
(org-element-footnote-reference-interpreter): Do not consider [1]-like
constructs as footnotes anymore.
* lisp/ox.el (org-export-expand-include-keyword):
(org-export--prepare-file-contents): Apply changes to footnotes.
* doc/org.texi (Footnotes): Remove references to plain footnotes.
* testing/lisp/test-org-element.el (test-org-element/footnote-reference-parser):
(test-org-element/footnote-reference-interpreter):
* testing/lisp/test-ox.el (test-org-export/expand-include):
(test-org-export/expand-macro):
(test-org-export/get-footnote-number):
(test-org-export/collect-footnote-definitions):
(test-org-export/footnotes):
(test-org-export/fuzzy-link): Update tests.
Since it is possible to refer unambiguously to a label without the "fn:"
prefix, the latter becomes part of the syntax and no longer part of the
label. In particular [fn:1] and [fn:label] are labelled, respectively,
"1" and "label".
* lisp/org-element.el (org-element-export-block-parser):
(org-element--current-element): Use new syntax for export blocks.
(org-element-block-name-alist): Remove.
* lisp/ox.el (org-export-register-backend):
(org-export-define-derived-backend): Remove :export-block keyword from
back-end's definition.
(org-export-expand-include-keyword): Use new syntax for include
keywords.
* lisp/ox-ascii.el (ascii):
* lisp/ox-beamer.el (beamer):
* lisp/ox-html.el (html):
* lisp/ox-latex.el (latex):
(org-latex-prefer-user-labels):
* lisp/ox-man.el (man):
* lisp/ox-md.el (md):
* lisp/ox-odt.el (odt):
* lisp/ox-texinfo.el (texinfo): Propagate changes to "ox.el".
* lisp/ob-core.el (org-babel-insert-result):
* lisp/ob-haskell.el (org-babel-haskell-export-to-lhs): Use new syntax.
* lisp/org.el (org-structure-template-alist): Use new syntax.
(org-protecting-blocks): Add "export" to protected blocks. Remove
back-ends blocks, which are now special blocks.
(org-fontify-meta-lines-and-blocks-1): Fontify correctly new blocks.
* lisp/org-footnote.el (org-footnote-forbidden-blocks):
* lisp/org-list.el (org-list-forbidden-blocks): Add "export" to
forbidden blocks. Remove back-ends blocks, which are now special
blocks.
* contrib/lisp/ox-extra.el: Update comments.
* testing/lisp/test-org-element.el (test-org-element/export-block-parser):
(test-org-element/export-block-interpreter):
* testing/lisp/test-ob.el (test-ob/org-babel-remove-result--results-html):
(test-ob/org-babel-remove-result--results-latex):
* testing/lisp/test-ox.el (test-org-export/expand-include): Update tests.
Export blocks are explicitly marked as such at the syntax level to
disambiguate their parsing from special blocks. The new syntax is
#+BEGIN_EXPORT backend
...
#+END_EXPORT
instead of
#+BEGIN_backend
...
#+END_backend
As a consequence, =INCLUDE= keywords syntax is modified, e.g.,
#+INCLUDE: "file.org" HTML
becomes
#+INCLUDE: "file.org" export html
* lisp/org-element.el (org-element--next-mode): Line following an
inlinetask can be a planning.
* testing/lisp/test-org-element.el (test-org-element/inlinetask-parser):
Update test.
Reported-by: Gregor Zattler <telegraph@gmx.net>
<http://permalink.gmane.org/gmane.emacs.orgmode/101958>
* lisp/org-element.el (org-element-link-parser): Call
`org-link-translation-function' if required.
(org-element-link-interpreter): Build link from type and path instead of
simply pasting raw value.
* lisp/org.el (org-translate-link): Call parser to extract proper path
and type.
* testing/lisp/test-org-element.el (test-org-element/link-interpreter):
Add test.
Reported-by: Sergei Nosov <sergei.nosov@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/100227>
* lisp/org-list.el (org-list-empty-line-terminates-plain-lists): Remove
variable.
(org-list-end-re):
(org-list-separating-blank-lines-number):
* lisp/org.el (org-blank-before-new-entry): Adapt to change above.
* testing/lisp/test-org-element.el (test-org-element/at-point):
* testing/lisp/test-org-list.el (test-org-list/move-item-down):
(test-org-list/move-item-up):
(test-org-list/insert-item): Update tests.
This variable is not useful: plain lists can always be terminated with
two blank lines.
* lisp/org-element.el (org-element-link-parser): Correctly parse links
with newline characters.
* lisp/org.el (org-make-link-regexps): Allow newline characters within
angle links.
(org-activate-angle-links): Support multiline angle links.
* testing/lisp/test-org-element.el (test-org-element/link-parser):
Update tests.
As a rule of thumb, any newline character and all surrounding
whitespaces are treated as a single space in a bracket link. They are
removed in angle links.
* lisp/org-element.el (org-element-table-parser): Properly parse an
Org table followed by a line starting with a "+".
* testing/lisp/test-org-element.el (test-org-element/context): Fix
test.
* lisp/org-element.el (org-element--cache-sync-requests): Remove a now
useless element from requests
(org-element--cache-submit-request): Apply change to sync request.
(org-element--cache-process-request): Apply change to sync requests.
Fix removal of orphaned elements, i.e., elements not affected by
a change, but with an ancestor that was.
* testing/lisp/test-org-element.el (test-org-element/cache): Add test.
Reported-by: Suvayu Ali <fatkasuvayu+linux@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/98260>
* lisp/org-element.el (org-element--set-regexps): Refine
`org-element-paragraph-separate'.
(org-element-paragraph-parser): Fix paragraph parsing when followed by
a dual keyword or an ill-formed latex environment.
* testing/lisp/test-org-element.el (test-org-element/paragraph-parser):
Add tests.
Reported-by: Jacob Gerlach <jacobgerlach@gmail.com>
<http://permalink.gmane.org/gmane.emacs.orgmode/97256>
* lisp/org-element.el (org-element--cache-for-removal): Be more
careful when altering a properties drawer.
* testing/lisp/test-org-element.el (test-org-element/cache): Add test.
* lisp/org-element.el (org-element-document-properties): Remove
variable.
(org-element-context): Ignore ex-"document keywords".
* testing/lisp/test-org-element.el (test-org-element/context): Remove
no longer applicable tests.
Parsed keywords are no longer hard-coded and can be set per export
back-end.
As a consequence, `org-element-context' no longer returns objects in
"TITLE", "DATE" or "AUTHOR" keywords. This makes sense as these
objects would only make sense during export, and only if the chosen
back-end actually uses them.
* lisp/org-element.el (org-element-line-break-parser): Tiny
refactoring.
(org-element-normalize-contents): Take into consideration line breaks
when removing common indentation. Small refactoring.
* testing/lisp/test-org-element.el (test-org-element/normalize-contents):
Add tests.
* lisp/org-element.el (org-element-object-containers): New variable
(org-element-interpret-data): Remove optional argument.
(org-element--interpret-data-1): Remove optional argument. Guess when
a given element or object has a pseudo type.
* lisp/ox-latex.el (org-latex-pseudo-objects): Remove variable.
(org-latex-table, org-latex--math-table): Apply signature change above.
* testing/lisp/test-org-element.el (test-org-element/interpret-data):
Update test.
* lisp/org-element.el (org-element-link-type-is-file): Remove.
(org-element-link-parser): Do not treat specially anymore "docview"
link type.
* testing/lisp/test-org-element.el (test-org-element/link-parser):
Update test.
* lisp/org-docview.el (org-docview-export): Fix path.
"docview" is a custom link type handled in its own library. There is
no reason to include it in core parser.
* testing/examples/babel.org: Change spaces to dashes in #+name
lines.
* testing/lisp/test-ob-exp.el (org-test-with-expanded-babel-code):
(ob-exp/evaluate-all-executables-in-order):
(ob-exp/exports-inline-code):
(ob-exp/exports-inline):
(ob-exp/exports-inline-code-double-eval):
(ob-exp/exports-inline-code-eval-code-once):
(ob-exp/exports-inline-code-double-eval-exports-both):
(ob-exp/use-case-of-reading-entry-properties):
(ob-exp/export-from-a-temp-buffer):
(ob-export/export-with-results-before-block):
(ob-export/export-under-commented-headline):
* testing/lisp/test-ob-lob.el (test-ob-lob/export-lob-lines):
* testing/lisp/test-ob.el (test-org-babel/inline-src_blk-default-results-replace-line-1):
(test-org-babel/inline-src_blk-default-results-replace-line-2):
(test-org-babel/inline-src_blk-manual-results-replace):
(test-org-babel/inline-src_blk-results-scalar):
(test-org-babel/inline-src_blk-results-verbatim):
(test-org-babel/inline-src_blk-preceded-punct-preceded-by-point):
* testing/lisp/test-ox.el (test-org-export/export-scope):
Bind ‘org-babel-inline-result-wrap’ and/or ‘org-export-babel-evaluate’
so tests work when users have customized these variables.
* testing/lisp/test-ob-exp.el (ob-export/export-with-results-before-block):
Add ‘org-trim’ where an extra newline was creeping in.
* testing/lisp/test-ob-lob.el (test-ob-lob/call-with-header-arguments):
Neutralize org-babel-insert-result, which was stomping on the buffer
contents and disrupting the test
* testing/lisp/test-org-element.el (test-org-element/block-switches):
Replace (should (and ...)) with multiple (should ...). This gives
more precise indications of what is wrong when a test fails.
(test-org-element/link-parser): Require org-docview.
* testing/lisp/test-org-timer.el (test-org-timer/other-timer-error):
Add the error type for ‘should-error’.
* testing/lisp/test-ox.el (test-org-export/set-title):
(test-org-export/handle-options):
(test-org-export/with-timestamps):
(test-org-export/comment-tree):
(test-org-export/handle-inlinetasks): Let
‘org-export-filter-body-functions’ and
‘org-export-filter-final-output-functions’ to nil where an empty
string is the expected result. ‘org-export-filter-apply-functions’
treats an empty string as special, and changes it to nil. This
creates test failures when the user has customized these variables.
* testing/org-test.el (org-test-with-temp-text): Let ‘org-mode-hook’
to nil.
(org-test-table-target-expect): Require ert since this function calls
some of its should* functions.
Nicolas Goaziou
Aaron Ecay