| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
very popular)
|
| |
|
| |
|
|
|
|
| |
This fix reverts the changes from #381 (i.e issue #4214 "htmlonly content appears in generated XML output (Origin: bugzilla 646002)", as XML is seen as an output format that can be further processed and therefore should contain all possible information.
|
|
|
|
|
| |
Made code analogous to HTML code.
Also the doxygen internal tests suffered from this problem.
|
|
|
|
|
|
|
|
| |
On Windows the syntax with the command and arguments in one string worked but on *nix (and Cygwin) it didn't.
- changed calls to Popen to split command (see also: https://docs.python.org/3/library/subprocess.html).
- explicitly specify files to check for xhtml
- changed call to xmllint for xhtml and docbook (due to stdout overflow in some cases, we are not really interested in the, formatted, output of xmllint).
- in the update part a `read` was left, should have been removed.
|
| |
|
|\
| |
| |
| | |
https://github.com/albert-github/doxygen into albert-github-feature/bug_html_s
|
| |
| |
| |
| |
| |
| | |
Adding the HTML `<s>` tag (https://www.w3schools.com/tags/tag_s.asp).
Adding 's' and correcting 'inc' to 'ins' in compound.xsd
|
| |
| |
| |
| |
| | |
In #7254 the tests were made for python 3 for the standard xml tests, now this is extended for the other possible formats.
The subprocess call gave some problems as it handles only 1 command and no redirection.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Test 65 gave some problems with python3:
```
Traceback (most recent call last):
File "D:/Programs/Doxygen/Doxygen-.git/doxygen/testing/runtests.py", line 487, in <module>
main()
File "D:/Programs/Doxygen/Doxygen-.git/doxygen/testing/runtests.py", line 484, in main
sys.exit(testManager.perform_tests())
File "D:/Programs/Doxygen/Doxygen-.git/doxygen/testing/runtests.py", line 388, in perform_tests
tester = Tester(self.args,test)
File "D:/Programs/Doxygen/Doxygen-.git/doxygen/testing/runtests.py", line 13, in __init__
self.config = self.get_config()
File "D:/Programs/Doxygen/Doxygen-.git/doxygen/testing/runtests.py", line 71, in get_config
for line in f.readlines():
File "D:\Programs\Python\Python37\lib\encodings\cp1252.py", line 23, in decode
return codecs.charmap_decode(input,self.errors,decoding_table)[0]
UnicodeDecodeError: 'charmap' codec can't decode byte 0x81 in position 236: character maps to <undefined>
```
this happened on multiple places and also with the `popen` command.
Created, analogous to `doc/translator.py`, special open functions so that the code works for Python 2 and Python 3.
|
|\ \
| | |
| | | |
Spelling corrections for testing directory
|
| | |
| | |
| | |
| | |
| | |
| | | |
Spelling corrections as found by codespell and in #561.
Most reported problems were already fixed, others are fixed here.
|
|/ / |
|
|/ |
|
|\
| |
| |
| | |
https://github.com/albert-github/doxygen into albert-github-fetaure/bug_html_ins_del_tag
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Github does not support the `<u>` tag for underlined text, in HTML there exists the tag `<ins>`. On https://www.w3schools.com/tags/tag_ins.asp is written:
The `<ins>` tag defines a text that has been inserted into a document.
Browsers will normally strike a line through deleted text and underline inserted text.
Most browsers will display the `<ins>` element with the following default values:
```
ins {
text-decoration: underline;
}
```
analogous there exists the tag `<del>` (https://www.w3schools.com/tags/tag_del.asp):
The `<del>` tag defines text that has been deleted from a document.
Browsers will normally strike a line through deleted text and underline inserted text.
Most browsers will display the `<del>` element with the following default values:
```
del {
text-decoration: line-through;
}
```
Definitions analogue to the underline and strike through tag the implementation for the other formats has been chosen.
|
|\ \
| | |
| | | |
regression #7061 Rename test file
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Due to a regression on #7061 travis showed message:
```
not ok 71 - [071_enum_in_anon_ns.cpp]: test that enum values in anonymous namespaces produce no warning
-------------------------------------
/home/travis/build/doxygen/doxygen/testing/071/namespace_a_namespace_1_1_0d0.xml absent
-------------------------------------
```
|
|/ / |
|
|\ \ |
|
| |\ \
| | | |
| | | | |
issue #7033 Syntax highlighting doesn't work when calling \dontinclude from Markdown files
|
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Markdown files
Don't use the file name extension of the including file but of the included file. In case of no included filename or no extension, fall back of the file extension of the including file.
|
|/ / / |
|
| | |
| | |
| | |
| | | |
In case special characters in a filename need escaping into something like: `_0` the `0` should not be followed by an uppercase character as this would be in contradiction wit the setting `CASE_SENSE_NAMES=NO`.
|
|/ /
| |
| |
| | |
Test 19 was incorrect, didn't show group description either (see make tests TEST_FLAGS="--keep --xhtml --id=19"
|
| | |
|
|/
|
|
|
|
| |
Create possibility to use CLANG_ASSISTED_PARSING, works only when doxygen has been compiled with "use_libclang"
Note at this moment the test suite gives an error, see issue #6948
|
|
|
|
| |
Added option `--xmlxsd` to check generated XML files against XSD in the doxygen tests suite.
|
|\
| |
| | |
Adjustment of xhtml1-transitional.dtd
|
| |
| |
| |
| | |
After small discussion conclusion was better to remove the obsolete `doxygen=`
|
| |
| |
| |
| |
| |
| |
| |
| | |
Adding attributes:
- doxygen
to be able to validate doxygen generated xhtml documents.
Analogous to #6681 but now for the area tag (found in the doxygen tag example).
|
|/ |
|
|\
| |
| | |
Double id for tooltips in XHTML possible.
|
| |
| |
| |
| | |
The tooltips (XHTML) should not be displayed per code fragment but for the entire (output) file as otherwise tooltips might be added multiple times resulting in double IDs.
|
|/
|
|
| |
Update test
|
| |
|
|\
| |
| |
| | |
https://github.com/albert-github/doxygen into albert-github-feature/bug_pre_markdown
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
According to the documentation:
Doxygen does not have this requirement, and will also process Markdown formatting inside such HTML blocks. The
only exception is <pre> blocks, which are passed untouched (handy for ASCII art).
Though in case the `</pre>` is the last statement in a non block type comment the markdown is processed. C has block type /* ... */ and /// is translated to this, Fortran, Python are line type comments.
Adding a `\n` at the when no `\n` is present at the end solves the problem.
|
| | |
|
|\ \ |
|
| |/
| |
| |
| | |
Test case.
|
|/
|
|
| |
in the XML output
|
|\
| |
| | |
issue #6733 invalid cite anchor id when using crossref
|
| |
| |
| |
| | |
Also test 050 was hit by the change of samlpe.bib
|
| |
| |
| |
| |
| |
| |
| | |
The crossreference possibility was not properly taken into account in bibtex conversion for other formats than LaTeX.
- doxygen.bst: use correct labels (i.e. add 'CITEREF_' in case of crossreferences to the giver name)
- cite.cpp: add the crossreferences to the citation dictionary (to overcome warning message).
- extending test 012
|
| |
| |
| |
| | |
length limitation
|
|\ \
| | |
| | |
| | | |
into mosra-extract-private-virtual
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
The classic article about virtuality from Herb Sutter [1] suggests that
virtual functions are always private and public class interface is never
virtual. Until now, it was not really possible to document these
functions in Doxygen:
* Enabling EXTRACT_PRIVATE would show all internals, not just virtual
functions, which is not wanted.
* Enabling HIDE_UNDOC_MEMBERS and HIDE_UNDOC_CLASSES would effectively
disable warnings about *all* undocumented members, which is not wanted.
The usual workaround was to put the members into protected scope just
for Doxygen:
#ifdef DOXYGEN_GENERATING_OUTPUT
protected:
#else
private:
#endif
/** @brief Documented private virtual function */
virtual doStuff();
The new EXTRACT_PRIVATE_VIRTUAL option makes these visible (and able to
be linked to), but shows them *only* if they are documented.
[1] http://www.gotw.ca/publications/mill18.htm
|
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | |
| | | |
Test 66 gives some warning messages bout not documented entities:
.../testing/066_property_initializer.cs:4: warning: Compound Class1 is not documented.
.../testing/066_property_initializer.cs:5: warning: Member Property1 (property) of class Class1 is not documented.
.../testing/066_property_initializer.cs:6: warning: Member Property2 (property) of class Class1 is not documented.
The entities have been documented.
|