ds000172 - dataset_description.json contains "un-braking space" (c2 a0) characters after lastnames

[yarikoptic]

which breaks stock json modules in python

[poldrack]

that is some really remarkable brittleness for a stock module…

On Aug 29, 2016, at 9:53 AM, Yaroslav Halchenko [email protected] wrote:

which breaks stock json modules in python
https://camo.githubusercontent.com/6ba2eb0e9c2520043e1c3c4d4011fb4bcbd95329/687474703a2f2f7777772e6f6e657275737369616e2e636f6d2f746d702f676b72656c6c53686f6f745f30382d32392d31365f3132353135352e706e67
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub #5, or mute the thread https://github.com/notifications/unsubscribe-auth/AA1KkJGbsRpgnA5WFISIvdJIOG3gORL-ks5qkw50gaJpZM4Jvq1b.

—
Russell A. Poldrack
Albert Ray Lang Professor of Psychology
Bldg. 420, Jordan Hall
Stanford University
Stanford, CA 94305

[email protected]
http://www.poldracklab.org/

[vsoch]

I ran into something like this yesterday... are there any web forms involved with getting this data? The HTML spec uses CR LF pairs:

http://www.w3.org/TR/html401/interact/forms.html#h-17.13.4.1

which means '/r/n' and those can be replaced easily when parsing the data before writing the json.

[vsoch]

You can also set strict=False when loading, or just replace the characters when you do open inside the json loads function!

[yarikoptic]

I doubt those characters are a part of json spec... imho it is unreasonable to demand json parsers to understand all those screwy utf8 symbols used as delimiters. I guess validator should also check those jsons more thoroughly. Later will file about at least one more ;-)

On August 29, 2016 12:57:29 PM EDT, Russ Poldrack [email protected] wrote:

that is some really remarkable brittleness for a stock module…

On Aug 29, 2016, at 9:53 AM, Yaroslav Halchenko
[email protected] wrote:

which breaks stock json modules in python

https://camo.githubusercontent.com/6ba2eb0e9c2520043e1c3c4d4011fb4bcbd95329/687474703a2f2f7777772e6f6e657275737369616e2e636f6d2f746d702f676b72656c6c53686f6f745f30382d32392d31365f3132353135352e706e67
—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#5, or mute the thread
https://github.com/notifications/unsubscribe-auth/AA1KkJGbsRpgnA5WFISIvdJIOG3gORL-ks5qkw50gaJpZM4Jvq1b.

—
Russell A. Poldrack
Albert Ray Lang Professor of Psychology
Bldg. 420, Jordan Hall
Stanford University
Stanford, CA 94305

[email protected]
http://www.poldracklab.org/

You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub:
#5 (comment)

[vsoch]

Excellent colors btw, is that the "Dreaming of McDonalds hamburger" terminal theme? +1! 🍟🍔

[yarikoptic]

More of "I was so cool when I was young" :-)
Thanks for the hint with ignoring option... Will check when get to the laptop

On August 29, 2016 2:03:18 PM EDT, Vanessa Sochat [email protected] wrote:

Excellent colors btw, is that the "Dreaming of McDonalds hamburger"
terminal theme? +1! 🍟🍔

You are receiving this because you authored the thread.
Reply to this email directly or view it on GitHub:
#5 (comment)

[yarikoptic]

@vsoch apparently strict=False wasn't enough. So will do fixups manually

[vsoch]

Yeah, I wound up just getting rid of them entirely before writing the file.

[jbwexler]

So would it be helpful for us to upload a new revision of this dataset that fixes this issue? Or is it easy enough to work around? Is it something we should consider with future datasets?

[yarikoptic]

I just worked around for now, so as to me, no rush ;-)