notebook -- spurious u0027's output

[williamstein]

In the notebook we have this, caused by _eval_cmd in worksheet.py:

%python 
2+2
print "'hi\'"
///

'hi\u0027

Component: notebook

Author: Willem Jan Palenstijn, Tim Dumol

Reviewer: Mitesh Patel, Tim Dumol

Merged: sagenb-0.8

Issue created by migration from https://trac.sagemath.org/ticket/3154

[TimDumol]

Attachment: trac_3154-spurious-u0027-output.patch.gz

Uses base64.b64en/decode instead.

[TimDumol]

Author: Tim Dumol

[wjp]

comment:2

I think using %r to handle escaping quotes would be cleaner than using intermediate base64. I'm attaching a new patch that does this.

[wjp]

Attachment: 3154_escaping_quotes.patch.gz

[qed777]

comment:3

Can we also use %r instead of base64-encoding in

• sage.misc.preparser.load_wrap
• sagenb.misc.support.automatic_name_filter
• worksheet.Worksheet.preparse

?

[qed777]

comment:4

The second patch causes two SageNB doctest failures:

File "/home/tmp/sagenb-0.5/src/sagenb/sagenb/notebook/worksheet.py", line 3695:
    sage: W.check_for_system_switching(c1.cleaned_input_text(), c1)
Expected:
    (True, u"print _support_.syseval(gap, ur'''SymmetricGroup(5)''', '...')")
Got:
    (True, u"print _support_.syseval(gap, u'SymmetricGroup(5)', '/home/.sage/temp/chopin/5101/dir_2.sagenb/home/sage/0/cells/1')")
**********************************************************************
File "/home/tmp/sagenb-0.5/src/sagenb/sagenb/notebook/worksheet.py", line 3721:
    sage: W.check_for_system_switching(c1.cleaned_input_text(), c1)
Expected:
    (True, u"print _support_.syseval(gap, ur'''SymmetricGroup(5)''', '...')")
Got:
    (True, "print _support_.syseval(gap, u'SymmetricGroup(5)', '/home/.sage/temp/chopin/5101/dir_2.sagenb/home/sage/0/cells/1')")

Does the latter reveal a [minor] problem with #7249?

[TimDumol]

comment:5

Replying to @qed777:

Can we also use %r instead of base64-encoding in

• sage.misc.preparser.load_wrap
• sagenb.misc.support.automatic_name_filter
• worksheet.Worksheet.preparse

?

I believe so. I'd rather that be put in a new ticket though.

The attached patch should solve the mentioned doctest problems. Can't see how they're related to #7249 though.

[TimDumol]

comment:6

Sorry, I forgot to attach the actual patch.

[TimDumol]

Fixes a few doctests and a unicode encoding issue.

[qed777]

Attachment: 3154_escaping_quotes.2.patch.gz

Rebase for minor "hunk" failure. Replaces previous.

[qed777]

Reviewer: Mitesh Patel

[qed777]

Changed author from Tim Dumol to Willem Jan Palenstijn, Tim Dumol

[qed777]

comment:7

Attachment: 3154_escaping_quotes.3.patch.gz

Nice work! V3 is just a rebase of V2.

[TimDumol]

Merged: sagenb-0.6

[qed777]

Rebased vs. SageNB 0.7.4. Replaces previous.

[qed777]

comment:9

Attachment: 3154_escaping_quotes.4.patch.gz

[qed777]

Work Issues: Rebase and fix tests

[qed777]

Changed merged from sagenb-0.6 to none

[qed777]

comment:10

V4 is rebased against SageNB 0.7.4 (cf. #8051), but now several doctests fail, at least one of which I can't investigate lucidly right now. I'll return to this soon.

[qed777]

comment:11

I've reopened this ticket because its not in SageNB 0.7 (cf. #8051). I mistakenly thought, per hg log, that I had merged it, but I really merged #4217, whose commit string refers to #3154.

[qed777]

Doctest fixes. Replaces all previous.

[qed777]

comment:12

Attachment: 3154_escaping_quotes.5.patch.gz

V5 is rebased for SageNB 0.7.4 and it includes several new doctest fixes. Can someone review my changes?

[qed777]

Changed work issues from Rebase and fix tests to none

[TimDumol]

comment:13

Doctests pass, no regressions noted.

[TimDumol]

Changed reviewer from Mitesh Patel to Mitesh Patel, Tim Dumol

[TimDumol]

Merged: sagenb-0.8