Render reports with styled notes containing subscript and strikethrough #1762
Conversation
| @@ -200,7 +204,6 @@ def escape(self, preformatted=False): | |||
| def find_tag_by_stag(self, s_tag): | |||
| """ | |||
| :param s_tag: object: assumed styledtexttag | |||
| :param s_tagvalue: None/int/str: value associated with the tag | |||
There was a problem hiding this comment.
There is no s_tagvalue parameter to the method, so I removed this comment.
|
@emyoulation once the fix is merged, we ought to add subscript and strikethrough markup to a note in the Gramps example database. Thanks. |
|
I've had a pedigree 'breadcrumb_list' item on the back-burner for quite a long time now. (Partly because the way Perplexity suggested for superscripts was an awkward substitution key method and not terribly cross-compatible.) Will this give a more flexible option? Is it viable to add SmallCaps styling too? That's the other common Genealogy report styling to distinquish surnames from other name parts. It would give Gramps a full complement of typical differentiating stylings for surnames: bold, underline, italic, uppercase. Small caps would be like the other stylings... it would disappear when pasted into a plain text editor. Upper case has to be hamd-converted. See |
Yes, I believe so and it would be best handled as a new feature request. This PR is a bug fix for an exception being thrown while generating reports with subscript & strikethrough being used in a Note. Previously @Nick-Hall implemented support for these styles in the Note editor with #1266 which worked well, and to fully integrate across gramps features, Smallcaps styling could follow the same pattern - implement support in the note editor, DocBackend and its implementations, and add the APIs (which remains a todo item for Gramps 5.3) to enable plugin and addon authors to use the styling as well. |
FYI I believe that is against the projects Howto: Contribute to Gramps
|
Yes, it is against the policy and a large number of open source projects are struggling with this and in most cases any person who attempts to contribute ai generated code is essentially attempting to taint the projects code base and in a number of cases those people are blocked from contributing 😢 which is not what you really want in an open source project. Even in this pr both the submitter and commenter are using ai's; the submitter does not mention which but the commenter has used at least [Perplexity.ai] see Gramps discussion (“Lumpers” and “Splitters” in Gramps, 2 drafts from Perplexity.ai) which in this touches on the user manual! So it is difficult to say what each of the ai's copyright material they trained on! Please note this is not a personal attack on either of the people involved here and more of comment on ai and contribution that probably needs to be discussed and sorted out, like maybe making an small ai model that only trains on Gramps code? |
|
@wroldwiedbwe @PQYPLZXHGF and @Nick-Hall The result will be that the existing defect |
|
Yes. Please remove the code that was written by the AI. We introduced this policy in April 2023 to avoid any potential future legal issues. As copyright law evolves to clarify issues relating to AI and code development, we may wish to revisit this discussion. |
|
Follow up discussion on Gramps forum |
Thanks. Filed feature request. |
Fixes #13417
With this change all reports should be able to render styled notes using subscript or strikethrough without throwing exceptions.
1. Add support for subscript and strikethough in DocBackend.
2. Implement support for subscript and strikethrough in:
a. CairoBackend
b. LatexBackend
i. Use package ulem for strikethrough support. ulem overrides rendering of emphasis, so specify normalel option to maintain the default.
ii. A bug in str_inc was being triggered during my testing. I used AI to fix the code and it seems to work. I haven't investigated whether this will change anything, but the code seems to work. Looking for review on the code change, ways to test, or other comments
c. HTMLBackend
d. ODFBackend
i. Add two new styles GSub and GStrikethrough to support subscript and strikethrough formatting, and implemented the two in odfdoc.py
Testing
1. Create a note that includes subscript and strikethrough, and attach to a person
2. Run any report which would include that note. Example: Complete Individual Report for that person
3. Render report in all formats and verify that (1) no exceptions are thrown and (2) output is rendered appropriately for the format. For Text and RTF, these styles will not have any effect, other formats including PDF, HTML, LaTeX, ODT, PS you will see the style applied.
Todo
1. Add methods similar to start[end]_superscript for subscript and strikethrough in each of the Doc classes so that plugins and addons can use the same functionality. For compatibility, this should be introduced in 5.3 so plugins can specify the version required.
2. Add similar API for italics (emphasize) since no API is available for that today. See thread on discourse: https://gramps.discourse.group/t/reports-normal-italic-and-bold-within-a-paragraph/5106/5
Code fix to str_incr() which was generated by AI has been reverted as this is explicitly disallowed by Gramps contribution rules.
hgohel
force-pushed
the
bug-13417-subscript-strikethough-support
branch
from
51c1b3b
to
c2afea9
Compare
Fixes #13417
With this change all reports should be able to render styled notes using subscript or strikethrough without throwing exceptions.
Use package ulem for strikethrough support. ulem overrides rendering of emphasis, so specified normalel option to maintain the default.
A bug instr_incwas being triggered during my testing. I used AI to fix the code and it seems to work. I haven't investigated whether this will change anything, but the code seems to work. Looking for review on the code change, ways to test, or other commentsUpdate: AI generated code reverted as it is explicitly forbidden by Gramps contribution rules. Bug #13418 filed to track this issue separately as it is not the target of the bug being addressed by PR.
Add two new styles GSub and GStrikethrough to support subscript and strikethrough formatting, and implemented the two in odfdoc.py
Testing
Todo