"sage -startuptime" should accept file names too

[mkoeppe]

(from #30410)

This is so that all of the tox environments (#30410, #30408) can be run on files and directories.

Example:

$ ./sage -startuptime src/sage/tensor/ src/sage/geometry/polyhedron/base.py

CC: @dimpase @tobiasdiez @fchapoton @jhpalmieri @tscrim @vbraun

Component: doctest framework

Author: Matthias Koeppe

Branch/Commit: 306c390

Reviewer: John Palmieri

Issue created by migration from https://trac.sagemath.org/ticket/30452

[mkoeppe]

Branch: u/mkoeppe/_sage__startuptime__should_accept_file_names_too

[mkoeppe]

Author: Matthias Koeppe

[mkoeppe]

New commits:

06c4b55

src/bin/sage-startuptime.py: Accept file/directory names too

[mkoeppe]

Commit: 06c4b55

[jhpalmieri]

comment:5

It's certainly true that ./sage --startuptime src/sage/homology/simplicial_set.py raises an error without this branch and doesn't raise an error with it, but with this branch it produces no output. What is it supposed to do?

By the way, it makes me a little uncomfortable to use the same loop variable twice like this:

for module_arg in cmdline_args:
    ...
    for module_arg in matching_modules:
        ...

[jhpalmieri]

comment:6

On the other hand, ./sage -startuptime src/sage/geometry/polyhedron/base.py does produce output. I'm confused.

[jhpalmieri]

comment:7

I would also recommend an explicit import os statement. I'm not sure why the script works without it. Edit: I guess it's being imported from sage.all, but it would be better to make it explicit at the top.

[mkoeppe]

comment:8

Without this ticket ./sage --startuptime sage.homology.simplicial_set gives

[]
Traceback (most recent call last):
  File "/Users/mkoeppe/s/sage/sage-rebasing/worktree-algebraic-2018-spring/src/bin/sage-startuptime.py", line 148, in <module>
    raise ValueError('"' + module_arg + '" does not uniquely determine Sage module.')
ValueError: "sage.homology.simplicial_set" does not uniquely determine Sage module.

Before working on this ticket, I had never used these scripts. I don't really know whether the above is expected, or a result of undetected bitrot over the years.

I think that it only works for modules that are loaded when sage.all is loaded.

[jhpalmieri]

comment:9

Is it worth printing a message if the file or module is not loaded when sage.all is loaded, rather than being completely silent?

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

25e1c5f	src/bin/sage-startuptime.py: import os
4256c9f	src/bin/sage-startuptime.py: Both with file and module argument, warn if there is no matching module; non-uniqueness is not an error

[sagetrac-git]

Changed commit from 06c4b55 to 4256c9f

[sagetrac-git]

Branch pushed to git repo; I updated commit sha1. New commits:

306c390

src/bin/sage-startuptime.py: Use different variable in nested loop

[sagetrac-git]

Changed commit from 4256c9f to 306c390

[mkoeppe]

comment:12

How about this version?

[jhpalmieri]

Reviewer: John Palmieri

[jhpalmieri]

comment:13

Great, thank you for the changes.

[mkoeppe]

comment:14

Thanks!

[vbraun]

Changed branch from u/mkoeppe/_sage__startuptime__should_accept_file_names_too to 306c390