New Port: textproc/smi Smi is a Simple Markup Interpreter / filter for simplified Markup dialect. smi can be fed text in Markdown, and return HTML output. smi can be fed HTML, and return the markup translated to entities. I use smi as a filter for devel/cgit to parse the README.md files, returning HTML output. I am also using it to markup wiki pages, for a git backed wiki. The use cases are limited only by your imagination. WWW: http://BSDforge.com/projects/textproc/smi/ PR: 196224 Submitted by: Chris Hutchinson <portmaster@bsdforge.com> Approved by: feld|delphij|pgollucci (mentor)
Details
Details
- Reviewers
pgollucci delphij feld - Commits
- rP396547: New Port: textproc/smi
Poudriere:
https://bugs.FreeBSD.org/bugzilla/show_bug.cgi?id=196224#c5
Portlint:
portlint -AC
looks fine.
Runtime:
The port appears to convert markdown to HTML as intended.
Diff Detail
Diff Detail
- Repository
- rP FreeBSD ports repository
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
| textproc/smi/Makefile | ||
|---|---|---|
| 7 ↗ | (On Diff #8490) | use ${PORTNAME} instead of smi -- less churn if the software is ever renamed |
| 23 ↗ | (On Diff #8490) | I opened up the config.mk. I personally would prefer to be a bit more specific on this REINPLACE in the event that somehow "cc" ends up elsewhere in the file (unlikely, but I'm paranoid). It also has "g" for global replace, but you really only intend to replace it once... get what I mean? Also, there's the following in config.mk: # paths
PREFIX = /usr/local
MANPREFIX = ${PREFIX}/share/man/usr/local shouldn't be hardcoded and neither should MANPREFIX as these are fully configurable in the ports tree. I'm curious to know if this thing would be perfectly happy with changing CC = to CC ?=, etc. If so, you could just do something like: @${REINPLACE_CMD} -e 's|^CC =|CC ?=|' -e 's|^PREFIX =|PREFIX ?=|' -e 's|^MANPREFIX =|MANPREFIX ?=|' ${WRKSRC}/config.mk |
Comment Actions
Much of this appeared to be a hold over from the original textproc/smu port (https://github.com/Gottox/smu) that this is based on. If this is approved, I'll push the suggestions over to the textproc/smu maintainer as they are valid there.
The upstream went to Github just before the PR was submitted in December 2014. Keeping all the legacy probably made sense then Chris said he has been working new features he intended to get in after this port was added. If that is the case, he can address them in the smi port code rather than needing to address this in ports.
| textproc/smi/Makefile | ||
|---|---|---|
| 24 ↗ | (On Diff #8621) | The ?= is a shell parse error on the port but I've addressed the intent of this. I've tested with PREFIX=/opt and CC=gcc48 in my make.conf and validated that this does the right thing. |