[fossil-users] README Maturity Model & CI/automated testing

2017-10-27 Thread Warren Young
Someone wrote up a document laying out opinions about how “mature” a given software project’s top-level README file is: https://github.com/LappleApple/feedmereadmes/blob/master/README-maturity-model.md It’s good food for thought, except insofar as it confuses “GitHub” with “the way

[fossil-users] Patch: tech_overview.wiki

2017-10-27 Thread Johan Kuuse
Hi, There are a few broken links in the Technical Overview Wiki page: Index: www/tech_overview.wiki == --- www/tech_overview.wiki +++ www/tech_overview.wiki @@ -53,11 +53,11 @@ checkout for a project and contains state

[fossil-users] Possible Markdown memleak

2017-10-27 Thread Johan Kuuse
Hi, I think the two blobs used for markdown input and output (rendered HTML) aren't free()'d properly. This patch only applies to test_markdown_render(), but the same behaviour may occur at other places where markdown_to_html() is called. BR, Johan Index: src/wiki.c

Re: [fossil-users] [fossil-dev] Possible Markdown memleak

2017-10-27 Thread Richard Hipp
On 10/27/17, Johan Kuuse wrote: > Hi, > > I think the two blobs used for markdown input and output (rendered > HTML) aren't free()'d properly. I'm sure your are correct about the leaks, though two points: (1) The proper way to fix them is to call blob_reset() on the Blob object,

Re: [fossil-users] [fossil-dev] Possible Markdown memleak

2017-10-27 Thread Richard Hipp
On 10/27/17, Richard Hipp wrote: > > (2) Memory leaks do not usually matter in Fossil, since each command > runs to completion and the process dies, leaving the OS to clean up > any memory. Only memory leaks inside of a loop make a difference. Another way to think of this: The

Re: [fossil-users] [fossil-dev] Possible Markdown memleak

2017-10-27 Thread Richard Hipp
On 10/27/17, Johan Kuuse wrote: >> > Obviously, this is not a big issue unless the Fossil web server > handles heavy traffic of big Markdown documents - which isn't its main > purpose anyway. :-) Even then, it isn't an issue, because the Fossil web server creates a new process to