Thanks Steve. I'll only address a couple of your specific issues inline. We can split the rest of the list if you'd like, but I think a lot of them are on the same page in the wiki (although multiple pages in the PDF) - let me know.
Cassandra On Thu, Sep 26, 2013 at 7:29 AM, Steve Rowe <sar...@gmail.com> wrote: > 0. All examples in the exported PDF have an extra blank line at the top. I > was able to eliminate these from this page > <https://cwiki.apache.org/confluence/pages/viewpage.action?pageId=32604227> > ("What is an analyzer?") by eliminating the newline between the initial {code > …} line and the first line of the examples. This doesn't have any apparent > effect on the layout of the page on the wiki, but the PDF export of that page > no longer has the extra blank lines. Any objections to switching all {code} > examples in the guide like this? CT: is it that horrible? There are dozens and dozens of code examples, and it will take a while for someone to fix all of them. Since I edit in wiki markup mode, I've always found it easier to add the line break so my eyes can find the samples faster. That said, ease of use for users is more important than my convenience, so if you think it's badly distracting, then it's worth trying to fix it. An alternative might be to try to change the CSS that produces the code examples - the problem is that the default styling for the PDF includes some padding, and then puts in the newline. Fiddling with the CSS is painful though - we can't see the interim HTML and it's essentially trial & error over & over. So, it's essentially one of two annoying choices: edit all the code examples by hand, or generate the PDF x-dozen times to maybe find out the CSS approach won't work. > > 1. Pg 2: The section links from the TOC all take you to the previous page, > rather than to the top of the page where the section starts. (Same behavior > on OS X Preview, and under Windows, on Firefox's built-in PDF viewer and on > Adobe Reader.) This looks like a general problem - see e.g. #34. CT: This is essentially a known problem (see my comment: https://issues.apache.org/jira/browse/SOLR-4886?focusedCommentId=13703660#comment-13703660, last bullet point). The way the PDF is created is that Confluence creates the entire document in an HTML page, which include bookmark tags right before the different heading levels. When the PDF is then generated, a rule is applied to insert a page-break before all h2 headings. That leaves the bookmark orphaned on the previous page. I have never found a solution to this problem - you can't edit the HTML and you don't have any control over where the bookmark tags in the HTML are put before the HTML is converted to PDF. The only solution is to never have page breaks, which I think severely diminishes readability. > > 2. Pg 68: Stray asterisks in the <analyzer> tags in the <fieldType> example > under "Analysis Phases", apparently to make the surrounded text bold (which > also didn't happen). CT: BTW, it never will - code examples are rendered verbatim, without any of the styling normally applied. > 43. Pg 106: Langauge-Specific Factories: Catalan, Danish, Irish and Romanian > are missing from the covered languages; Catalan and Irish should include > ElisionFilterFactory in their examples - there are articles lists in Lucene's > {Catalan,Irish}Analyzer. CT: A general note about the languages and examples - there used to be examples that were incorrect so were removed so that might account for some of the gaps. There's an open issue you'll want to look at before diving in: https://issues.apache.org/jira/browse/SOLR-5031. --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org For additional commands, e-mail: dev-h...@lucene.apache.org
