4 ms·

The punchline here is a help target that digs through your makefiles looking for comments: help: @grep -E '^[a-zA-Z_-]+:.?## .$$' $(MAKEFILE_LIST) | sort

The punchline here is a help target that digs through your makefiles looking for comments: help: @grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort | awk 'BEGIN {FS = ":.*?## "}; {printf "\033[36m%-30s\033[0m %s\n", $$1, $$2}' I stumbled over .*? since ? usually means optional, but turns out it means lazy in this context. The $$ would be makefile escaping. Dropping the grep and changing the regex slightly, I just appended this to an overengineered makefile: # Help idea derived from https://marmelab.com/blog/2016/02/29/auto-documented-makefile.html # Prints the help-text from `target: ## help-text`, slightly reformatted and sorted .PHONY: help help: ## Write this help awk 'BEGIN {FS = ":.*#+"}; /^[a-zA-Z_*.-]+:.*## .*$$/ {printf "%-30s %s\n", $$1, $$2}' $(MAKEFILE_LIST) | sort Very nice idea, thanks for sharing it!

konfekt 3y ago
Which make version does the latter command use? For GNU make 4.2.1 on Linux, I hade more luck with something along the lines of @awk '/^[a-zA-Z_-]+:[^#]*## .*$$/ {match($$0, /## (.*)$$/, arr); printf "\033[36m%-30s\033[0m %s\n", $$1, substr($$0, RSTART+3)}' $(MAKEFILE_LIST)
- JonChesterfield 3y ago
  Looks like 4.3 but I don't think it matters - awk vs gawk/nawk might be significant though, gawk 5.2 on the machine I ran this on. The match with substr is interesting. It's more complicated than setting the field separator to something like :|#+ but should mean : in the help text works. For something one only writes and debugs once, probably better to do the complicated thing that always works. gawk will write the groups to an array, that's possibly more legible (and slower? should be slower than the leading non-capture //) @gawk 'match($$0, /^([a-zA-Z_*.-]+):.*## (.*)$$/, arr) {printf " %-30s %s\n", arr[1], arr[2]}' $(MAKEFILE_LIST) | sort
ceving 3y ago
If you can give up regexp greediness sed+column can do it: sed -rn 's/^([^:]+):.*[ ]##[ ](.+)/\1:\2/p' $(MAKEFILE_LIST) | column -ts: -l2 | sort
- JonChesterfield 3y ago
  Sed confuses me more than awk but you're right. That would also remove the only use of awk in my makefile (sed is there already for hacking around spaces in filenames). Whitespace padding output in sed is probably horrible, column looks simpler than printf via bash or trying to use make's $info.
  - defrost 3y ago
    Whilst acknowledging that "Confused by SED" has overlap with "have used SED for 40 years and have three books purely on SED" I can recommend https://www.grymoire.com/Unix/Sed.html https://www.grymoire.com/Unix/Sed.html as a reference some might occasionally swear by. Anyhow, sed is a marvelous utility. Unfortunately, most people never learn its real power. The language is very simple, but the documentation is terrible. The Solaris on-line manual pages for sed are five pages long, and two of those pages describe the 34 different errors you can get. A program that spends as much space documenting the errors as it does documenting the language has a serious learning curve.
    - JonChesterfield 3y ago
      Appreciated, reading through it. I suspect the majority of the sed experience can be attributed to it using "posix regular expressions" by default. It was about a decade after first discovering sed that I realised passing -E was really important. It is difficult for newcomers to guess that "extended regular expressions" refers to the barely-usable subset of "regular expressions" and "posix regular expressions" are terrible in comparison to either. edit: alright, yes, one can program in that. Sed can recurse. .PHONY: help3 help3: sed -nE 's/^([a-zA-Z_*.-]+):.*## (.*)$$/\1 :\2/ p' \ $(MAKEFILE_LIST) | \ sed -E -e ':again s/^([^:]{1,16})[:]([^:]+)$$/\1 :\2/ ' -e 't again ' |\ sed -E 's/^([^ ]*)([ ]*):(.*)$$/\1:\2\3/' |\ sort The first invocation filters out the lines of interest, second one space pads to 16. That works by putting the colon before the help text and repeatedly inserting a space before the colon until there are at least sixteen non-colon characters in the first group. Composing the -n/p combination with act on everything is a stumbling block for merging the multiple invocations together but I expect it to be solvable.
rcarmo 3y ago
I have been doing this for a long while now, and it's great

The punchline here is a help target that digs through your makefiles looking for comments: help: @grep -E '^[a-zA-Z_-]+:.*?## .*$$' $(MAKEFILE_LIST) | sort

konfekt 3y ago

Which make version does the latter command use? For GNU make 4.2.1 on Linux, I hade more luck with something along the lines of @awk '/^[a-zA-Z_-]+:[^#]*## .*$$/ {match($$0, /## (.*)$$/, arr); printf "\033[36m%-30s\033[0m %s\n", $$1, substr($$0, RSTART+3)}' $(MAKEFILE_LIST)

JonChesterfield 3y ago
Looks like 4.3 but I don't think it matters - awk vs gawk/nawk might be significant though, gawk 5.2 on the machine I ran this on. The match with substr is interesting. It's more complicated than setting the field separator to something like :|#+ but should mean : in the help text works. For something one only writes and debugs once, probably better to do the complicated thing that always works. gawk will write the groups to an array, that's possibly more legible (and slower? should be slower than the leading non-capture //) @gawk 'match($$0, /^([a-zA-Z_*.-]+):.*## (.*)$$/, arr) {printf " %-30s %s\n", arr[1], arr[2]}' $(MAKEFILE_LIST) | sort

ceving 3y ago

If you can give up regexp greediness sed+column can do it: sed -rn 's/^([^:]+):.*[ ]##[ ](.+)/\1:\2/p' $(MAKEFILE_LIST) | column -ts: -l2 | sort

JonChesterfield 3y ago
Sed confuses me more than awk but you're right. That would also remove the only use of awk in my makefile (sed is there already for hacking around spaces in filenames). Whitespace padding output in sed is probably horrible, column looks simpler than printf via bash or trying to use make's $info.
- defrost 3y ago
  Whilst acknowledging that "Confused by SED" has overlap with "have used SED for 40 years and have three books purely on SED" I can recommend https://www.grymoire.com/Unix/Sed.html https://www.grymoire.com/Unix/Sed.html as a reference some might occasionally swear by. Anyhow, sed is a marvelous utility. Unfortunately, most people never learn its real power. The language is very simple, but the documentation is terrible. The Solaris on-line manual pages for sed are five pages long, and two of those pages describe the 34 different errors you can get. A program that spends as much space documenting the errors as it does documenting the language has a serious learning curve.
  - JonChesterfield 3y ago
    Appreciated, reading through it. I suspect the majority of the sed experience can be attributed to it using "posix regular expressions" by default. It was about a decade after first discovering sed that I realised passing -E was really important. It is difficult for newcomers to guess that "extended regular expressions" refers to the barely-usable subset of "regular expressions" and "posix regular expressions" are terrible in comparison to either. edit: alright, yes, one can program in that. Sed can recurse. .PHONY: help3 help3: sed -nE 's/^([a-zA-Z_*.-]+):.*## (.*)$$/\1 :\2/ p' \ $(MAKEFILE_LIST) | \ sed -E -e ':again s/^([^:]{1,16})[:]([^:]+)$$/\1 :\2/ ' -e 't again ' |\ sed -E 's/^([^ ]*)([ ]*):(.*)$$/\1:\2\3/' |\ sort The first invocation filters out the lines of interest, second one space pads to 16. That works by putting the colon before the help text and repeatedly inserting a space before the colon until there are at least sixteen non-colon characters in the first group. Composing the -n/p combination with act on everything is a stumbling block for merging the multiple invocations together but I expect it to be solvable.

rcarmo 3y ago

I have been doing this for a long while now, and it's great