Using Jinja2 condition against a string

[xcke]

Hi All,

Maybe a user error, but if I use a jinja2 if condition something like:

{% if site_csv.Rack_Type =='RACK1' %}
bla bla
{% endif %}

I receive error during running:
jinja2.exceptions.TemplateSyntaxError: unexpected char '”' at 101224

Same setup with an integer seems to be working. I tried different chars like ",',` none of them working to identify string.

[elapouya]

Are you using MS Word 97 or something else ? 2017-06-20 14:33 GMT+02:00 xcke <notifications@github.com>:

…

Hi All, Maybe a user error, but if I use a jinja2 if condition something like: {% if site_csv.Rack_Type =='RACK1' %} bla bla {% endif %} I receive error during running: jinja2.exceptions.TemplateSyntaxError: unexpected char '”' at 101224 Same setup with an integer seems to be working. I tried different chars like ",',` none of them working to identify string. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#85>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA2N7G-Cgp0G18fyLsqyxqKtEoR-PJ22ks5sF7wSgaJpZM4N_hCi> .

[xcke]

Ms Word 2013, the file is in compatibility mode.

[xcke]

Same issue with the standard 2013 file format.

[xcke]

Workaround is to copy paste the Jinja2 code from Pycharm. In that case the ' " ' will be "fine".

[elapouya]

I am a little bit surprised, why pycharm (that I use) is changing something ?? 2017-06-20 16:03 GMT+02:00 xcke <notifications@github.com>:

…

Workaround is to copy paste the Jinja2 code from Pycharm. In that case the ' " ' will be "fine". — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA2N7H9I4WHpl4iAOR54hcx2DpYsAKH4ks5sF9E8gaJpZM4N_hCi> .

[xcke]

I think this has to be something about char coding of Ms. Word, but I tried different fonts for the Jinja2 syntax. And if you copy paste the Jinja2 code from Pycharm, in that case in Word it will be kept using the right format?

[elapouya]

I guess so... 2017-06-20 16:30 GMT+02:00 xcke <notifications@github.com>:

…

I think this has to be something about char coding of Ms. Word, but I tried different fonts for the Jinja2 syntax. And if you copy paste the Jinja2 code from Pycharm, in that case in Word it will be kept using the right format? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AA2N7E4fQyejkPAI3EBEgjCCglvTRbLvks5sF9eQgaJpZM4N_hCi> .

[elapouya]

Could you send me a minimal docx template and test python file so I can test my side ?

[HandwovenBox]

I believe I experienced the same error. When I typed in quotes for the condition, Word put in the slanted quotes automatically, which led to the error: ”

I manually replaced both slanted quotes with straight quotes " and then it worked.

I am attaching two .docx files that are identical except for the "not_working" file has the slanted quotes.

Conditional_Test.docx
Conditional_Test_not_working.docx

Edit: This is for the most recent version-- Word 2016

[xcke]

Good news:) HandwovenBox <notifications@github.com> (időpont: 2017. aug. 9., Sze, 10:57) ezt írta:

I believe I experienced the same error. When I typed in quotes for the condition, Word put in the slanted quotes automatically, which led to the error: ” I manually replaced both slanted quotes with straight quotes " and then it worked. I am attaching two .docx files that are identical except for the "not_working" file has the slanted quotes. Conditional_Test.docx <https://github.com/elapouya/python-docx-template/files/1210881/Conditional_Test.docx> Conditional_Test_not_working.docx <https://github.com/elapouya/python-docx-template/files/1210882/Conditional_Test_not_working.docx> — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#85 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/APl08rMEBcNyuPRelibEfANyQVgeVZhUks5sWXRvgaJpZM4N_hCi> .

-- *Gabor Kis-Hegedus* IP Network Engineer, CCIE #43897 | e: gabor.kh@gmail.com | w: internetworking.hu [image: Facebook] <https://www.facebook.com/xckee> [image: Twitter] <https://twitter.com/xcke> [image: LinkedIn] <https://www.linkedin.com/in/gaborkh> [image: Skype] xckesrac

[HandwovenBox]

So here's how to fix this issue (maybe--it worked for my very simple test). Add the following code at line 47

    def patch_xml(self,src_xml):
        #remove all “curly” quotes (single and double) found within {% %} and replace with straight
        def stripquotes(m):
            return re.sub(r'[“”‘’]+','\"',m.group(0),flags=re.DOTALL)
        src_xml = re.sub(r'(?<=\{\%).*?(?=\%\})',stripquotes,src_xml,flags=re.DOTALL)

I'm interested in hearing whether that works for anybody else.

[elapouya]

Good idea : I will try...

[milesgranger]

@HandwovenBox This worked delightfully, thank you! Would be nice if this was added. 👍

[ptr0x1]

This is an issue with Word IMO, as it switches the standard "-character to the unicode “ version which is obviously not supported as a valid tag. The same problem could come up if it switches the dash character '-' to the unicode long dash '–'.