[ 
https://issues.apache.org/jira/browse/DOXIA-554?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15930963#comment-15930963
 ] 

Vladimir Schneider commented on DOXIA-554:
------------------------------------------

Modified tests and flexmark-java options to not output extra formatting EOLs, 
closer to pegdown's unformatted output. Now all passing.

Next hurdle:

{code}
[INFO] --- clirr-maven-plugin:2.7:check (default) @ doxia-module-markdown ---
[INFO] Comparing to version: 1.7
[ERROR] Unable to locate enclosing class 
org.apache.maven.doxia.module.markdown.MarkdownParser for nested class 
org.apache.maven.doxia.module.markdown.MarkdownParser$PegDownHtmlParser
[ERROR] 8001: 
org.apache.maven.doxia.module.markdown.MarkdownToDoxiaHtmlSerializer: Class 
org.apache.maven.doxia.module.markdown.MarkdownToDoxiaHtmlSerializer removed
{code}

The MarkdownToDoxiaHtmlSerializer should have never been public in the original 
markdown module. It is purely an implementation detail. I also left the 
PegDownHtmlParser class name as is to avoid more errors, although it should be 
named MarkdownHtmlParser to make it independent of the parsing library.

Is there something I am missing in the configuration to allow an API breaking 
change to be allowed for the module?

The alternative of leaving the dead class and dependency on pegdown does not 
seem like the right solution.

My fork was updated with the changes to date. 


> Parsing time for Markdown documents can take very long and hang site 
> generation
> -------------------------------------------------------------------------------
>
>                 Key: DOXIA-554
>                 URL: https://issues.apache.org/jira/browse/DOXIA-554
>             Project: Maven Doxia
>          Issue Type: Bug
>          Components: Module - Markdown
>    Affects Versions: 1.7
>            Reporter: Michael Benz
>         Attachments: maven-pom-sample-pegdown-performance.zip
>
>
> The parsing time for Markdown documents can take very long and hang site 
> generation when e.g. long tables are being generated.
> The author of pegdown has marked the pegdown project deprecated since 
> 2016-12-14 [pegdown.org|https://github.com/sirthias/pegdown/] and advises to 
> switch to [flexmark-java|https://github.com/vsch/flexmark-java].
> {quote}
> The project is essentially unmaintained with tickets piling up and crucial 
> bugs not being fixed.
> pegdown's parsing performance isn't great. In some cases of pathological 
> input runtime can even become exponential, which means that the parser either 
> appears to "hang" completely or abort processing after a time-out.
> {quote}
> Since the parsing timeout was increased in DOXIA-498 it is now possible to 
> "hang" the site creation with a longer table like the one in this example.
> In case this sample is rendered using version 3.3 of the maven site



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Reply via email to