PEP 536: Final Grammar for Literal String Interpolation #155

flying-sheep · 2016-12-10T14:13:05Z

Hi, I announced some time ago that I’d pick this up, now I’ve finally managed to write it!

I hope the general structure, phrasing, formatting and so on are OK, this is my first PEP.

the-knights-who-say-ni · 2016-12-10T14:13:06Z

Hello, and thanks for your contribution!

Unfortunately we couldn't find an account corresponding to your GitHub username at bugs.python.org (b.p.o). If you don't already have an account at b.p.o, please create one and make sure to add your GitHub username. If you do already have an account at b.p.o then please go there and under "Your Details" add your GitHub username.

And in case you haven't already, please make sure to sign the PSF contributor agreement (CLA); we can't legally look at your contribution until you have signed the CLA.

Once you have done everything that's needed, please reply here and someone will verify everything is in order.

flying-sheep · 2016-12-10T14:18:56Z

I signed the CLA and put my github name!

I’d also love if #1 could be fixed. would make it easy to provide a “rendered view” link like it’s common for rust RFCs

brettcannon · 2016-12-10T20:26:00Z

Please wrap the lines at 70 or 79 characters (I haven't looked at the content).

flying-sheep · 2016-12-11T21:09:10Z

done! some code that i copied from PEP 498 needs to be longer.

brettcannon · 2016-12-12T19:38:48Z

pep-0XXX.txt

+Version: $Revision$
+Last-Modified: $Date$
+Author: Philipp Angerer <[email protected]>
+Status: Pending


Could you change this to Draft so Travis testing will pass?

brettcannon · 2016-12-12T21:04:57Z

Looks like Travis is still not happy yet.

flying-sheep · 2016-12-12T21:07:31Z

on it. apparently you have to give it a number anyway. maybe it would make sense to provide a way to get a number before a PEP or so.

berkerpeksag

The .rst extension doesn't work with python.org yet. Could you change it to .txt please?

berkerpeksag · 2016-12-12T21:06:18Z

pep-0999.rst

@@ -0,0 +1,175 @@
+PEP: 999


Is there a specific reason to use 999? I think we can use the next available number: 536 :)

PEP 12 says i have to be given one, not just pick the next one, but it seems reasonable. maybe we should fix this in PEP 12 and fix conflicts if people try to use the same number.

PEP 12 also says we can use rst already…

PEP 12 also says we can use rst already…

That's my fault. I thought I can finish my work by now, but due to some personal issues I couldn't find the time to work on it yet.

Don't worry and I hope everything works out for you!

berkerpeksag · 2016-12-12T21:06:36Z

pep-0999.rst

+Created: 20-Sep-2016
+Python-Version: 3.7
+Post-History: 20-Sep-2016
+Resolution: None


This is optional.

what do you mean? can i just leave out the “Resolution” line?

There is no point to keep an optional field now (and I don't think using None as a value makes any sense) We can always add it back when your PEP is discussed on python-ideas :)

berkerpeksag · 2016-12-12T21:07:14Z

pep-0999.rst

+Content-Type: text/x-rst
+Created: 20-Sep-2016
+Python-Version: 3.7
+Post-History: 20-Sep-2016


You probably want to update this date :)

Rosuav · 2016-12-12T21:13:55Z

PEP 12 probably needs some updates for the new GitHub workflow. Having a contributor pick a PEP number is unlikely to have conflicts (it's not common to see two new PEPs proposed simultaneously), and as soon as one's merged, the other would show a merge conflict in git. So it's safe.

flying-sheep · 2016-12-12T21:19:59Z

@berkerpeksag all done

berkerpeksag · 2016-12-12T22:17:45Z

pep-536.txt

+it is easy to extend syntax highlighters to correctly parse
+and display f-literals:
+
+.. raw:: html


python.org (django-markitup to be exact) ignores the raw directive for security reasons. Can you change this to a plain code block?

@sadly no. the whole point is to show this specific syntax highlighting.

are you sure python.org does this? i don’t see markitup in use here.

if it does though: PEPs are trusted. we can simply set raw_enabled to True. or we do it and use bleach to sanitize the HTML.

I think it would be more interesting to actually include the diff of what you had to change to the syntax highlighter to make it work than the actual output as the critical aspect of your claim is it's easy to update highlighters, not that it works in the end (we can just take your word on that).

As for removing the protection, I don't think we should change how python.org functions just for this one thing (we've gone 16 years and hundreds of PEPs without allowing raw HTML and this specific example is not critical to the PEP itself anyway).

I think it would be more interesting to actually include the diff of what you had to change to the syntax highlighter to make it work

good idea, i’ll add that part!

than the actual output

seeing is believing! i really want to include that part for people who have only seen string-colored code parts until now instead of something like this!

I don't think we should change how python.org functions just for this one thing

actually @berkerpeksag was mistaken and we don’t have to change a thing. generate_pep_pages.py says:

Run this command AFTER normal RST -> HTML PEP transformation from the PEP
repository has happened. This works on the HTML files created during that
process.

with “works”, it means “runs get_pep_page” on the paths of the generated PEP HTML fragments…

…and pep2html.py happily embeds my raw HTML

Good catch! I just tried this branch in my local python.org setup and it worked fine. I forgot we only store HTML outputs in the database. Thanks!

The only thing we need is to rename the file name to pep-0536.txt (add a leading zero like the other PEPs)

I haven't fully read the PEP yet so I'll let Brett or Chris do the proofreading and the merging.

whoops! done, the zero’s there.

@brettcannon i added a paragraph about syntax highlighter implementation

Mariatta · 2016-12-12T22:24:02Z

pep-536.txt

+Rationale
+=========
+
+.. https://mail.python.org/pipermail/python-ideas/2016-August/041727.html


This link should be at the References section, and add the thread title here (Let’s make escaping in f-literals impossible)

As mentioned in PEP 12 :)
https://www.python.org/dev/peps/pep-0012/#hyperlinks

ATM it’s just a comment, but i can add it as link if you want.

brettcannon · 2016-12-13T23:52:48Z

pep-0536.txt

+by giving f-literals a regular grammar without exceptions
+and implementing it using dedicated parse code.
+
+.. note:: put preceding paragraph into “Rationale”?


Should probably decide if you're going to follow through with this comment or simply remove it now as there's no need to wait on feedback for a decision.

brettcannon · 2016-12-13T23:54:52Z

pep-0536.txt

+Familiarity also plays a role: Arbitrary nesting of expressions
+without expansion of escape sequences is available in every single
+other language employing a string interpolation method that uses
+expressions instead of just variable names. [#]_


Do give names to your references, else it's easy to get the sequence wrong if you add a reference later on.

brettcannon · 2016-12-13T23:55:23Z

pep-0536.txt

+Python challenge”
+#. Syntax highlighters are good in parsing nested grammar, but not
+in recognizing escape sequences.  ECMAScript 2016 (JavaScript) allows
+escape sequences in its identifiers [#]_ and the author knows of no


Please name your references as its easy for the order to change if you add a reference later.

brettcannon · 2016-12-13T23:57:22Z

pep-0536.txt

+in expression portions of single line strings.
+
+.. note:: Is lifting of the restrictions sufficient,
+   or should we specify a more complete grammar?


An actual diff of Python's Grammar/Grammar file would be best.

could you point me to another PEP where this is done so i have a reference?

I don't know of a specific one off the top of my head. You're going to have to do this for the implementation anyway and so you can just embed the diff to the grammar file.

brettcannon · 2016-12-13T23:57:52Z

pep-0536.txt

+Reference Implementation
+========================
+
+TBD


Is an implementation being worked on?

not yet. isn’t the usual procedure to wait for feedback first?

The usual procedure is to ask on python-ideas if the idea seems viable, and then if it does then you do a draft PEP to python-ideas, and then if it holds up you do a rough implementation and send it along with the PEP to python-dev because that's when the question of maintenance comes into play.

OK. in this case people already confirmed that they think this is a good idea, but i never coded anything in C. so where do i turn to for guidance?

Either you ask for help upfront when you post your PEP to python-ideas and end up with a co-author or you hope that python-dev thinks your idea is so good that a core dev agrees to do it for you (but that's not very common).

brettcannon · 2016-12-14T18:49:33Z

I just realized I don't remember seeing this PEP sent to python-ideas and I forgot to ask that. So has this gone to python-ideas first for discussion prior to opening this PR?

flying-sheep · 2016-12-14T18:56:54Z

no, but the idea was born in a thread there, and all arguments already came up

not in comprehensive PEP form though

brettcannon · 2016-12-14T19:49:24Z

Since you need the PEP to ask for help I've gone ahead and accepted the PEP. Before you send it to python-ideas, @flying-sheep , please fork the repository and have people send changes directly to you. Once the discussion is finished on python-ideas you can send us another PR to incorporate the accumulated changes from the python-ideas discussion.

flying-sheep · 2016-12-15T14:24:51Z

OK, thank you! will do.

i’ll tell them to comment on https://github.com/flying-sheep/peps/blob/master/pep-0536.txt

the-knights-who-say-ni added the CLA not signed label Dec 10, 2016

flying-sheep changed the title ~~Added f-literals PEP draft~~ PEP ???: Final Grammar for Literal String Interpolation Dec 10, 2016

the-knights-who-say-ni added CLA signed and removed CLA not signed labels Dec 11, 2016

brettcannon requested changes Dec 12, 2016

View reviewed changes

brettcannon approved these changes Dec 12, 2016

View reviewed changes

berkerpeksag requested changes Dec 12, 2016

View reviewed changes

flying-sheep changed the title ~~PEP ???: Final Grammar for Literal String Interpolation~~ PEP 536: Final Grammar for Literal String Interpolation Dec 12, 2016

berkerpeksag reviewed Dec 12, 2016

View reviewed changes

Mariatta reviewed Dec 12, 2016

View reviewed changes

berkerpeksag approved these changes Dec 13, 2016

View reviewed changes

brettcannon requested changes Dec 13, 2016

View reviewed changes

Added f-literals PEP draft

a475b32

brettcannon approved these changes Dec 14, 2016

View reviewed changes

brettcannon merged commit 64eebdf into python:master Dec 14, 2016

flying-sheep deleted the f-literals branch December 15, 2016 14:22

flying-sheep mentioned this pull request Dec 25, 2016

Start recommending curly quotes or start using automatic quote conversion #165

Closed

PEP 536: Final Grammar for Literal String Interpolation #155

PEP 536: Final Grammar for Literal String Interpolation #155

Conversation

flying-sheep commented Dec 10, 2016 • edited Loading

the-knights-who-say-ni commented Dec 10, 2016

flying-sheep commented Dec 10, 2016 • edited Loading

brettcannon commented Dec 10, 2016

flying-sheep commented Dec 11, 2016

Choose a reason for hiding this comment

Choose a reason for hiding this comment

brettcannon commented Dec 12, 2016

flying-sheep commented Dec 12, 2016

berkerpeksag left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Rosuav commented Dec 12, 2016

flying-sheep commented Dec 12, 2016

Choose a reason for hiding this comment

flying-sheep Dec 12, 2016 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

flying-sheep Dec 13, 2016 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

flying-sheep Dec 14, 2016 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

brettcannon commented Dec 14, 2016

flying-sheep commented Dec 14, 2016

brettcannon commented Dec 14, 2016

flying-sheep commented Dec 15, 2016

flying-sheep commented Dec 10, 2016 •

edited

Loading

flying-sheep commented Dec 10, 2016 •

edited

Loading

flying-sheep Dec 12, 2016 •

edited

Loading

flying-sheep Dec 13, 2016 •

edited

Loading

flying-sheep Dec 14, 2016 •

edited

Loading