Thanks Blazer and Co-authors for this nice program.
It's the best desicion in this area, I've ever seen in web.
I've been used this (and previous) release for a while, and found a bunch of glitches:
1. "Queue" still works wrong in the cases, where user must enter some values, e.g. "Numeric Prefix" Margins. So, only first file is processed well, and all others got some mess up, because one option works only once (for one numeric prefix margin, for example). It works good only for non-optional tools, such as "Case Tools". And the best desicion here, I guess, is to remove all these each-file-user-input stuff.
So, If I want to Numeric Prefix 20 dictionaries, I should just input margins only once, at the very beginning.
2. "Horizontal Join" is unstable. For example I start combining one Dic with 2 other, and as a result I get a Dic which has only results untill J-letter, and not further. So, somethings interrupts the full process.
3. "Combinator" is unstable. This problem concerns non-english charset, e.g. Russian.
This is similar to point 2. something interrupts full process. Similar issue is with sorting of russian dictionaries, which works untill 80% of file, and then crushes the program. I tried to use UTF-8 but it didn't help.
4. "Case tools" don't work for non-english alphabets, e.g. Russian. So, I cannot uppercase such words, and perhaps to lowercase them to. This, however, can be resolved by "Direct Remapping", which, thank Tzeench and Blazer, works fine.
5. "Custom charset prefix/sufiix" is also unstable. This is similar ot point 2. and 3. Well, this works good for small dictionaries, but something interrupts the process for bigger ones. I haven't notice the size threshold in megabytes, though. I can try to track this issue if you like.
6. "Generator" crushes program, when there is no extension of file. So, for example, file z1 with no .txt proceeded this little bug.
Also, there are some whishes I would like this program to have.
1. Remembering of the last folder. Well this is not handy - to start each time from ULM root folder, when you work with lots of folders, containing different groupped types of dictionaries. The best approach is, I guess, to track last 10 (or whatever) paths which user used, similar as Total Commander do it. (Ok, there is some trackng of paths inside ULM, but it tracks only paths to files, not folders, which is not handy.)
2. More detailed descriptions. Well, laconicism is a good thing, but not for every theme, I suppose.
That is why, I haven't used some options yet, because I don't quite understand the idea. For example, I am still not sure how to use this program for parsing any text and transform it to dictionary. The best desicion is chm-file, I guess, or some simple example in the description for every tool.
3. Static charset adding. So, such tools as "Generator" contain default charsets, such as: Numeric, Alpha, Symbols, etc. This could be handy to load from the list some user-defined charsets, for quick access.
4. Uppercasing a definite number of letters per item. For example, I wish to generate dictionary with first 2 or 3 letters uppercased, and others are lowercased. By the way, the reverse operation would be great too, so, If I want uppercase last 3 letters of word, I would uppercase them all at first step, and then lowercase the definite number of first letters at the second step.
5. Numeric Calendar Prefix : ) The idea, is to have option to generate only those numbers which are dates, months or years. For example, I want to create a dictionary, for all dates from 2000 to 2005 with all full dates. E.g. Admin > Admin02052005, Admin03052005, etc. By the way, defining the format for the date with custom separator is also a good idea.
OK, that's enough for one post, I guess. Btw, Sorry for my pretty uperfect English.
Thanks again, for such great soft.
It's the best desicion in this area, I've ever seen in web.
I've been used this (and previous) release for a while, and found a bunch of glitches:
1. "Queue" still works wrong in the cases, where user must enter some values, e.g. "Numeric Prefix" Margins. So, only first file is processed well, and all others got some mess up, because one option works only once (for one numeric prefix margin, for example). It works good only for non-optional tools, such as "Case Tools". And the best desicion here, I guess, is to remove all these each-file-user-input stuff.
So, If I want to Numeric Prefix 20 dictionaries, I should just input margins only once, at the very beginning.
2. "Horizontal Join" is unstable. For example I start combining one Dic with 2 other, and as a result I get a Dic which has only results untill J-letter, and not further. So, somethings interrupts the full process.
3. "Combinator" is unstable. This problem concerns non-english charset, e.g. Russian.
This is similar to point 2. something interrupts full process. Similar issue is with sorting of russian dictionaries, which works untill 80% of file, and then crushes the program. I tried to use UTF-8 but it didn't help.
4. "Case tools" don't work for non-english alphabets, e.g. Russian. So, I cannot uppercase such words, and perhaps to lowercase them to. This, however, can be resolved by "Direct Remapping", which, thank Tzeench and Blazer, works fine.
5. "Custom charset prefix/sufiix" is also unstable. This is similar ot point 2. and 3. Well, this works good for small dictionaries, but something interrupts the process for bigger ones. I haven't notice the size threshold in megabytes, though. I can try to track this issue if you like.
6. "Generator" crushes program, when there is no extension of file. So, for example, file z1 with no .txt proceeded this little bug.
Also, there are some whishes I would like this program to have.
1. Remembering of the last folder. Well this is not handy - to start each time from ULM root folder, when you work with lots of folders, containing different groupped types of dictionaries. The best approach is, I guess, to track last 10 (or whatever) paths which user used, similar as Total Commander do it. (Ok, there is some trackng of paths inside ULM, but it tracks only paths to files, not folders, which is not handy.)
2. More detailed descriptions. Well, laconicism is a good thing, but not for every theme, I suppose.
That is why, I haven't used some options yet, because I don't quite understand the idea. For example, I am still not sure how to use this program for parsing any text and transform it to dictionary. The best desicion is chm-file, I guess, or some simple example in the description for every tool.
3. Static charset adding. So, such tools as "Generator" contain default charsets, such as: Numeric, Alpha, Symbols, etc. This could be handy to load from the list some user-defined charsets, for quick access.
4. Uppercasing a definite number of letters per item. For example, I wish to generate dictionary with first 2 or 3 letters uppercased, and others are lowercased. By the way, the reverse operation would be great too, so, If I want uppercase last 3 letters of word, I would uppercase them all at first step, and then lowercase the definite number of first letters at the second step.
5. Numeric Calendar Prefix : ) The idea, is to have option to generate only those numbers which are dates, months or years. For example, I want to create a dictionary, for all dates from 2000 to 2005 with all full dates. E.g. Admin > Admin02052005, Admin03052005, etc. By the way, defining the format for the date with custom separator is also a good idea.
OK, that's enough for one post, I guess. Btw, Sorry for my pretty uperfect English.
Thanks again, for such great soft.