CSV module, DictReader problem (bug?)

[Jeff Blaine]

It's been a year or so since I written Python code, so maybe
I am just doing something really dumb, but...

Documentation
=============

class DictReader(csvf ile[,fieldnames=Non e,
[,restkey=None[, restval=None[, dialect='excel'
[, *args, **kwds]]]]])


Create an object which operates like a regular reader
but maps the information read into a dict whose keys
are given by the optional fieldnames parameter. If the
fieldnames parameter is omitted, the values in the
first row of the csvfile will be used as the fieldnames.

Code
====

import csv

r = csv.DictReader( 'C:\Temp\Book1. csv')
print r.next()
# EOF

Output
======

{'C': ':'}

[Fredrik Lundh]

Re: CSV module, DictReader problem (bug?)

Jeff Blaine wrote:

It's been a year or so since I written Python code, so maybe
I am just doing something really dumb, but...
>
Documentation
=============
>
class DictReader(csvf ile[,fieldnames=Non e,
[,restkey=None[, restval=None[, dialect='excel'
[, *args, **kwds]]]]])
>
>
Create an object which operates like a regular reader
but maps the information read into a dict whose keys
are given by the optional fieldnames parameter. If the
fieldnames parameter is omitted, the values in the
first row of the csvfile will be used as the fieldnames.
>
Code
====
>
import csv
>
r = csv.DictReader( 'C:\Temp\Book1. csv')
print r.next()
# EOF

here's the documentation for the regular reader, from Python 2.5:

reader(...)
csv_reader = reader(iterable [, dialect='excel']
[optional keyword args])
for row in csv_reader:
process(row)

The "iterable" argument can be any object that returns a line
of input for each iteration, such as a file object or a list.
...

so the reader is simply looping over the characters in your filename. try

r = csv.DictReader( open('C:\Temp\B ook1.csv'))

instead.

</F>

[Jeff Blaine]

Re: CSV module, DictReader problem (bug?)

I see what's wrong. Me. Wow am I ever rusty.

Jeff Blaine wrote:

It's been a year or so since I written Python code, so maybe
I am just doing something really dumb, but...
>
Documentation
=============
>
class DictReader(csvf ile[,fieldnames=Non e,
[,restkey=None[, restval=None[, dialect='excel'
[, *args, **kwds]]]]])
>
>
Create an object which operates like a regular reader
but maps the information read into a dict whose keys
are given by the optional fieldnames parameter. If the
fieldnames parameter is omitted, the values in the
first row of the csvfile will be used as the fieldnames.
>
Code
====
>
import csv
>
r = csv.DictReader( 'C:\Temp\Book1. csv')
print r.next()
# EOF
>
Output
======
>
{'C': ':'}

[John Machin]

Re: CSV module, DictReader problem (bug?)

Jeff Blaine wrote:

It's been a year or so since I written Python code, so maybe
I am just doing something really dumb, but...
>
Documentation
=============
>
class DictReader(csvf ile[,fieldnames=Non e,
[,restkey=None[, restval=None[, dialect='excel'
[, *args, **kwds]]]]])
>
>
Create an object which operates like a regular reader
but maps the information read into a dict whose keys
are given by the optional fieldnames parameter. If the
fieldnames parameter is omitted, the values in the
first row of the csvfile will be used as the fieldnames.
>
Code
====
>
import csv
>
r = csv.DictReader( 'C:\Temp\Book1. csv')

Problem 1:

"""csvfile can be any object which supports the iterator protocol and
returns a string each time its next method is called -- file objects
and list objects are both suitable. If csvfile is a file object, it
must be opened with the 'b' flag on platforms where that makes a
difference."""

So, open the file, so that the next() method returns the next chunk of
content, not the next byte in the name of the file.

Note that the arg is called "csvfile", not "csvfilenam e".

Problem 2: [OK in this instance, but that's like saying you have taken
one step in a minefield and are still alive] backslashes and Windows
file names:

If you were to write 'c:\temp\book1. csv', it would blow up ... because
\t -tab and \b -backspace. Get into the habit of *always* using raw
strings r'C:\Temp\Book1 .csv' for Windows file names (and re patterns).
You could use double backslashing 'C:\\Temp\\Book 1.csv' but it's
uglier.

print r.next()
# EOF
>
Output
======
>
{'C': ':'}

HTH,
John

[Tom Plunket]

Re: CSV module, DictReader problem (bug?)

John Machin wrote:

If you were to write 'c:\temp\book1. csv', it would blow up ... because
\t -tab and \b -backspace. Get into the habit of *always* using raw
strings r'C:\Temp\Book1 .csv' for Windows file names (and re patterns).
You could use double backslashing 'C:\\Temp\\Book 1.csv' but it's
uglier.

....alternative ly you can just use 'unix slashes', e.g.
'c:/temp/book1.csv', since those work just fine 'cause the Windows
APIs deal with them properly.
-tom!

[John Machin]

Re: CSV module, DictReader problem (bug?)


Tom Plunket wrote:

John Machin wrote:
>

If you were to write 'c:\temp\book1. csv', it would blow up ... because
\t -tab and \b -backspace. Get into the habit of *always* using raw
strings r'C:\Temp\Book1 .csv' for Windows file names (and re patterns).
You could use double backslashing 'C:\\Temp\\Book 1.csv' but it's
uglier.

>
...alternativel y you can just use 'unix slashes', e.g.
'c:/temp/book1.csv', since those work just fine 'cause the Windows
APIs deal with them properly.

Not all APIs do the right thing. If you fire up the cmd.exe shell and
feed it slashes as path separators, it barfs. Example:
C:\junk>dir c:/junk/*.bar
Invalid switch - "junk".
Hence the advice to use rawstrings with backslashes -- they work under
all circumstances.

[skip@pobox.com]

Re: CSV module, DictReader problem (bug?)

>...alternative ly you can just use 'unix slashes', e.g.
>'c:/temp/book1.csv', since those work just fine 'cause the Windows
>APIs deal with them properly.

JohnNot all APIs do the right thing. If you fire up the cmd.exe shell
Johnand feed it slashes as path separators, it barfs. Example:
John C:\junk>dir c:/junk/*.bar
John Invalid switch - "junk".
JohnHence the advice to use rawstrings with backslashes -- they work
Johnunder all circumstances.

I think he means "the Windows APIs" within a Python program.

Skip

[Steve Holden]

Re: CSV module, DictReader problem (bug?)

John Machin wrote:

Tom Plunket wrote:
>

>>John Machin wrote:
>>
>>

>>>If you were to write 'c:\temp\book1. csv', it would blow up ... because
>>>\t -tab and \b -backspace. Get into the habit of *always* using raw
>>>strings r'C:\Temp\Book1 .csv' for Windows file names (and re patterns).
>>>You could use double backslashing 'C:\\Temp\\Book 1.csv' but it's
>>>uglier.

>>
>>...alternativ ely you can just use 'unix slashes', e.g.
>>'c:/temp/book1.csv', since those work just fine 'cause the Windows
>>APIs deal with them properly.

>
>
Not all APIs do the right thing. If you fire up the cmd.exe shell and
feed it slashes as path separators, it barfs. Example:
C:\junk>dir c:/junk/*.bar
Invalid switch - "junk".
Hence the advice to use rawstrings with backslashes -- they work under
all circumstances.
>

The command shell is not an API.

regards
Steve
--
Steve Holden +44 150 684 7255 +1 800 494 3119
Holden Web LLC/Ltd http://www.holdenweb.com
Skype: holdenweb http://holdenweb.blogspot.com
Recent Ramblings http://del.icio.us/steve.holden

[John Machin]

Re: CSV module, DictReader problem (bug?)

On 2/11/2006 2:38 PM, skip@pobox.com wrote:

>...alternative ly you can just use 'unix slashes', e.g.
>'c:/temp/book1.csv', since those work just fine 'cause the Windows
>APIs deal with them properly.

>
JohnNot all APIs do the right thing. If you fire up the cmd.exe shell
Johnand feed it slashes as path separators, it barfs. Example:
John C:\junk>dir c:/junk/*.bar
John Invalid switch - "junk".
JohnHence the advice to use rawstrings with backslashes -- they work
Johnunder all circumstances.
>
I think he means "the Windows APIs" within a Python program.
>
Skip
>

I too think he meant that. I left the mental gymnastics of wrapping what
I wrote into an os.system call as an exercise for the reader.

| >>import os
| >>os.system("di r c:/junk/*.bar")
| Invalid switch - "junk".
| 1
| >>os.system(r"d ir c:\junk\*.bar")
| [snip]
| 02/11/2006 01:21 PM 8 foo.bar
| [snip]
| 0
| >>>

Cheers,
John

[Fredrik Lundh]

Re: CSV module, DictReader problem (bug?)

John Machin wrote:

Not all APIs do the right thing. If you fire up the cmd.exe shell and
feed it slashes as path separators, it barfs. Example:
C:\junk>dir c:/junk/*.bar
Invalid switch - "junk".
Hence the advice to use rawstrings with backslashes -- they work under
all circumstances.

we went through this a couple of days ago; all Windows API:s are
*documented* to accept backward or forward slashes, the command shell is
*documented* to only accept backward slashes.

if you're wrapping some cmd.exe command in an internal API, it's usually
easier to call "os.path.normpa th" the last thing you do before you call
"os.system" , than to get all the backslashes right in your code.

also see:

http://www.effbot.org/pyfaq/why-can-t-raw-strings-r-strings-end-with-a-backslash.htm

</F>

[Fredrik Lundh]

Re: CSV module, DictReader problem (bug?)

Fredrik Lundh wrote:

if you're wrapping some cmd.exe command in an internal API, it's usually
easier to call "os.path.normpa th" the last thing you do before you call
"os.system" , than to get all the backslashes right in your code.
>
also see:
>
http://www.effbot.org/pyfaq/why-can-...-backslash.htm

and as just I pointed out in a comment on that page, getting the slashes
right doesn't help you with spaces in filenames. to write reliable code
for os.system, you want something like:

import os.path
import subprocess

def mysystem(comman d, *files):
files = map(os.path.nor mpath, files)
files = subprocess.list 2cmdline(files)
return os.system(comma nd + " " + files)

mysystem("more" , "/program files/subversion/readme.txt")

but then you might as well use subprocess.call , of course.

</F>