Shell Programming and Scripting

Hello,
I have a dictionary of over 400,000 words with the following structure

Code:

source=target

The database contains single words as well as phrases. To train the data, I need only mappings with out a space i.e. where both source and target do not have any space in between.
I use Ultraedit as my editor and have been using the following regex in Unix to identify a blank space

Code:

^[^ ]+$

Since the database is too large, the editor runs out of memory and cannot store all instances to the clipboard.
Am giving below a small sample text

Code:

هيراآباد=हीरा आबाद
واسڪوڊيگاما=वास्कोडीगामा
کانسواءِ=खांसवाइ/खा सिवाइ
آوازنکي=आवाज़नखे
سانآهي=सान आहे
سڏبوآهي=सॾबो आहे
شڪارڪرڻ=शकार करण
ٺاهيندوآهي=ठाहींदो आहे
ٻولينجو=ॿोलीनजो
ٻولينجي=ॿोलीनजे
ڪنديآهي=कंदी आहे
گئسنجو=गैसन जो
ماموغلام=मामूग़ुलाम
زاهدچانڊيو=ज़ाहिद चांडियो
عطرڪمار=अतुरकुमार
غلاممحي=ग़ुलाममही
گلشيرڪوريجو=गुलशेर कोरीजो
زيرحراست=ज़ीर हिरासत

The script should identify only those entries without a space on either side and store them in a separate file, as in the sample output below

Code:

واسڪوڊيگاما=वास्कोडीगामा
آوازنکي=आवाज़नखे
ٻولينجو=ॿोलीनजो
ٻولينجي=ॿोलीनजे
ماموغلام=मामूग़ुलाम
عطرڪمار=अतुरकुमार
غلاممحي=ग़ुलाममही

A perl or awk script would help. I work in a windows environment.
Many thanks.

Have a look at the -v option of grep:

Code:

grep -vF ' ' your_file

True. Still, as a measure of safety i would rule out trailing or leading spaces:

Code:

sed -n '/^[[:blank]]*//;s/[[:blank:]]*$//;/ /!p' > /result/file

I hope this helps.

bakunin

For instance using grep:

Code:

grep -v '[^ ] [^ ]' your_file

Many thanks for all your kind help. My broadband connectivity was down all day and hence the delay. All the solutions worked. I had ensured that my data had no trailing spaces so the issue of trailing spaces does not arise but is is nice to have a solution which ensures that trailing spaces are handled.
Thanks once again.