MOR with (un)til, (be)cause, etc.

[Leonid Spektor]

Hi Stephen,

	I just tried MOR on Mac Unix and it worked as expected. MOR also worked correctly on Mac GUI and Win PC. I looked at the source code to see if there was something unusual there and I didn't see anything. Finally, I stepped through source code that removes parentheses from words with a debugger and again I could see anything that would produce the result that you are getting. It is very strange that in your case only the first and the last word is deformed and the middle word "(un)til" get processed correctly. Have you tried the "-fwritable-strings" option with CFLAGS?

	First, I would like you to start emailing to me directly, instead of to chibolts. Second, I would like you to email to me the test file that you are using and the script of the MOR output, including the command line, from the Unix terminal window. I guess you could just highlight the text, copy it and paste it into email. If I don't find anything wrong with your input file, then I will consider installing Ubuntu on one of our machines here and see if I can replicate your result. The Ubuntu web site says that they are up to version 12.10 or 12.04 LTS now. So, that is the version that I'll be testing.

Leonid.



On Feb 11, 2013, at 15:37, Stephen Wilson wrote:

> Hi Leonid,
> 
> Thanks for your reply. I just downloaded the latest version of clan (V 08-Feb-2013 11:00), as well as the mor grammar (from http://childes.psy.cmu.edu/morgrams/), and recompiled. Unfortunately I got the same result as before:
> 
> *PAR:	(a)bout (un)til (th)em .
> %mor:	?|abutt conj|until^prep|until ?|theem .
> 
> I am using clan on Ubuntu 10.04. I compile with:
> CC = g++
> and
> CFLAGS = -DUNX -Wno-deprecated -Wno-deprecated-declarations
> 
> The first line of the make file mentions using the -m32 option, but that gives an error, so I don't do that.
> 
> Could you please let me know if you're able to replicate the strange behavior on unix clan, or if I might be doing something wrong when compiling or anything else?
> 
> Thanks a lot,
> 
> best,
> Stephen
> 
> 
> On Wednesday, February 6, 2013 3:20:03 PM UTC-7, Spektor, Leonid: CMU wrote:
> Stephen,
> 
> 	I just tested the latest version of CLAN on both Mac and Windows PC and after running MOR on your sample tier below I get the following correct result:
> 
> *PAR:	(a)bout (un)til (th)em .
> %mor:	prep|about^adv|about prep|until^conj|until pro:obj|them .
> 
> Please, make sure that you have the latest version of CLAN and MOR grammar. If you do have the latest versions, then please try to quit CLAN and restart it. Maybe CLAN got into bad state. If after using the latest CLAN/grammar and restarting CLAN, you still get this problem, then please email to me the data file and command line that produced bad result.
> 
> 
> Leonid.
> 
> 
> 
> On Feb 6, 2013, at 16:45 , Stephen Wilson wrote:
> 
>> MOR does not seem to handle "noncompletion" of words as expected, e.g. (un)til, (be)cause.
>> 
>> According to the CHAT manual: "CLAN will treat items that are coded in this way as full words. For programs such as FREQ, the parentheses will essentially be ignored and (be)cause will be treated as if it were because."
>> 
>> But does this not apply to MOR? I get this as my parse:
>> 
>> *PAR:	(a)bout (un)til (th)em .
>> %mor:	?|abutt conj|until^prep|until ?|theem .
>> 
>> thanks,
>> Stephen
>> 
>> -- 
>> You received this message because you are subscribed to the Google Groups "chibolts" group.
>> To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+u... at googlegroups.com.
>> To post to this group, send email to chib... at googlegroups.com.
>> To view this discussion on the web visit https://groups.google.com/d/msg/chibolts/-/_z9nvzG6BMwJ.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>  
>>  
> 
> 
> -- 
> You received this message because you are subscribed to the Google Groups "chibolts" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
> To post to this group, send email to chibolts at googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msg/chibolts/-/dQhsYnhK6hcJ.
> For more options, visit https://groups.google.com/groups/opt_out.
>  
>  

-- 
You received this message because you are subscribed to the Google Groups "chibolts" group.
To unsubscribe from this group and stop receiving emails from it, send an email to chibolts+unsubscribe at googlegroups.com.
To post to this group, send email to chibolts at googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20130211/0ec3385a/attachment.htm>