-
Notifications
You must be signed in to change notification settings - Fork 126
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
wrap new att lists in <ul>
(where they were not so wrapped)
#641
Conversation
sydb
commented
Oct 28, 2023
- template showAttClasses is called from 6 different places (all in common_tagdocs);
- generated debugging code for each of those places;
- found which 2 were causing problem (li element as child of td instead of child of ul);
- wrapped them in ul element.
- Note: deliberately left debugging PIs in place, but they should maybe be removed before merging into dev.
* template showAttClasses is called from 6 different places (all in common_tagdocs); * generated debugging code for each of those places; * found which 2 were causing problem (li element as child of td instead of child of ul); * wrapped them in ul element. * Note: deliberately left debugging PIs in place, but they should maybe be removed before merging into dev.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I'm looking for any trouble with @sydb 's repair in the output Guidelines. I see no problems with specs in:
[x] - Epub (inspected several element specs)
[x] - Guidelines PDF build
[x] - Guidelines HTML
Locally, the following tests pass:
[x] - TEI repo (no diffs from expected results)
Stylesheets Test 1 does not pass, breaks on
make[1]: *** [Makefile:583: test-oddity] Error 1
make[1]: Leaving directory '/stylesheet/Test'
make: *** [Makefile:106: test] Error 2
Stylesheets Test 2 passes on my Mac.
It's failing for me ONLY in Docker (and I suspect a Docker error?) on
- build.xml: 550
- build_docx.xml: 32, 43, and 64
Summarizing how things went by the end of the day on Saturday 10/29 (EST). @sydb notes (from Slack): Every test in Test/ now passes except test-epub, for which 13 of the output files get the error “This file should declare in the OPF the property: scripted”. In order to get the XSLT to run, the top of tei-to-epub3.xsl looks like: <xsl:import href="../html/html.xsl"/>
<xsl:import href="../odds/teiodds.xsl"/> <!-- only included for $autoGlobal param, which ../odds/classatts.xsl relies on -->
<xsl:import href="../odds/classatts.xsl"/> <!-- only included for ATTCLASSES key -->
<xsl:import href="../epub/epub-common.xsl"/>
<xsl:import href="../epub/epub-preflight.xsl"/>
<xsl:import href="../html/html_oddprocessing.xsl"/> <!-- only included for schemaOut template, which ../odds/teiodds.xsl seems to need --> Syd notes: We need to figure out the correct file for the "schemaOut" template (there are 5 of them, all different). Perhaps this is better than using <xsl:variable name="key" select="@key"/>
<!-- note: following line does not use "key('ATTCLASSES', @key)" because ATTCLASSES is not defined in this or any imported file, and importing it requires importing other files, too. -->
<xsl:variable name="thisClassSpec" select="//tei:classSpec[ @ident eq $key]" as="element(tei:classSpec)?"/> |
I've now tested this (adding the all of those
I believe this is caused by an import, which is adding all of the HTML javascript into the epubs but shouldn't be there — or, at least, isn't there in Adding the following to the bottom of <!--Add empty javascript hook to remove JS-->
<xsl:template name="javascriptHook"/> The suggestion of replacing |
Implemented change I suggested above and which @joeytakeda thought was the way to go in 7ed5111. Has passed test of building the Guidelines. Running Stylesheet tests now. |
After removing In Test/ the file actual-results/test15.odd.html has 4 extraneous empty Note |
@sydb: Do we know what is it that's causing them? At the moment, it produces an output that has (unsurprisingly) empty additional list items (below screenshot of output of I wonder if we should reverse course and take the brute-force approach of adding the imports and the empty javascript hook? While this is certainly more overhead, it does allow both test suites to pass (and I don't imagine would cause problems in other transformations, unlike changes to |
No, I do not really know what is causing them. As for going the “adding the imports and the empty javascript hook” route, do we know what this test file looks like with that system? |
I found the bug, I think: the |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This appears to be working. Ready for merge as soon as the tests all complete.