chat2xml and pauses

[Tim Boardman]

Hello
I have been trying to convert a set of .cha files to .xml using the
chat2xml program (version alpha-53) and seem to be having a problem
with pauses (#).
For example the file
http://www.splloc.soton.ac.uk/spllocdb/transcripts/Discussion/D50D57MJA13.cha
once it has had @Coder: changed to @Transcriber: returns the errors
*** File "D50D57MJA13.cha": line 18, column 17: lexer: skipping
nonviable character '#' in state 48
*** File "D50D57MJA13.cha": line 27, column 48: lexer: skipping
nonviable character '#' in state 48
*** File "D50D57MJA13.cha": line 27, column 58: lexer: skipping
nonviable character '#' in state 48
*** File "D50D57MJA13.cha": line 61, column 23: lexer: skipping
nonviable character '#' in state 48
*** File "D50D57MJA13.cha": line 62, column 47: lexer: skipping
nonviable character '#' in state 48
and a blank xml file.

I have previously converted this file to XML without difficulty
(http://www.splloc.soton.ac.uk/spllocdb/xml/Discussion/D50D57MJA13.xml)

I can remove all of the # symbols from the .cha file but this would
seem to remove information from the file and offer no benefits.
What is the best action to take?

--
Thanks
Tim Boardman
SPLLOC & FLLOC Projects
University of Southampton

[Leonid Spektor]

Tim,

Character '#' is no longer legal CHAT format. If you want to use the
latest version of chat2xml, then you need to update CLAN to the latest
version and to use the following commands to convert old pause symbols '#'
to new ones. Here are two commands:

chstring +cpauses.cut *.cha
ren -f *.str.cex *.cha

If the above doesn't work try this one command:

chstring +1 +cpauses.cut *.cha

Before you run either of the above commands please make sure that CLAN's lib
is set to CLAN/lib folder.

Leonid.

[Sue Blackwell]

Dear Leonid and all,

Has the "columns" command been definitively abolished? I did
ask about this some time back and recall that it was being looked
into.

It was such a useful command for (a) producing handouts for teaching
purposes showing the different speakers, and (b) giving students an
easy command to try out in the introductory seminar on using CLAN.

Any chance of reviving it in the new version of CLAN?

many thanks,

Sue

--
Dr. Sue Blackwell
Department of English,
The University of Birmingham,
Edgbaston,
BIRMINGHAM B15 2TT

Phone: +44 - 0121-414-3219
Fax: +44 - 0121-414-5668

e-mail: S.A.Bl...@bham.ac.uk

Official University page:
http://www.english.bham.ac.uk/who/blackwell.htm
Unofficial University Home Page: http://web.bham.ac.uk/sue_blackwell
Personal, Trade Union and Political Pages: http://www.sue.be

[Brian MacWhinney]

Dear Sue,
I am afraid that we can't commit Leonid's time to this now. If
we resurrect COLUMNS,
we would want to do so in a more general way. However, even in that
case, the formatting issues involved in making
COLUMNS output fit into the right "squares" are pretty serious. I can
imagine a highly restricted simple
form of COLUMNS that allows for only two speakers and no dependent
tiers. But, given the limitations
this would impose, I wonder if it would not end up being misleading.

-- Brian

[Ann Peters]

I just want to go on record that I, too, have found Columns very useful in
my work. I understand the formatting problems, and also the priority
problems, but being able to skim down a *column* of one person's speech,
while still being able to access what else is going on, has been extremely
useful.
ann

****************************
Dr. Ann M. Peters, Professor Emeritus
Co-Graduate Chair http://www.ling.hawaii.edu/
Department of Linguistics
University of Hawai`i email: a...@hawaii.edu
1890 East West Road, Rm 569 phone: 808 956-3241
Honolulu, HI 96822 fax: 808 956-9166
http://www.ling.hawaii.edu/faculty/ann/

[Sue Blackwell]

Thanks for reply Brian. I understand the pressures on Leonid's time
and energies, but it's a pity. Even a simplified version would be
better than nothing. I hope you will keep this one on the back burner
for when Leonid has more time. Perhaps I will keep an older version of
CLAN on one of my laptops so that I can generate Columns output when I
need to.

Sue B

Brian MacWhinney wrote:
> Dear Sue,
> I am afraid that we can't commit Leonid's time to this now. If
> we resurrect COLUMNS,
> we would want to do so in a more general way. However, even in that
> case, the formatting issues involved in making
> COLUMNS output fit into the right "squares" are pretty serious. I can
> imagine a highly restricted simple
> form of COLUMNS that allows for only two speakers and no dependent
> tiers. But, given the limitations
> this would impose, I wonder if it would not end up being misleading.
>
> -- Brian