Re: [WebDNA] remove html comments
This WebDNA talk-list message is from 2009
It keeps the original formatting.
numero = 101818
interpreted = N
texte = Owkaye Go wrote: >> (just under 800 lines) and used: >> and it successfully parsed out all the > > I think you say is correct and original answer > wrong. I will test and report my results. Can you > explain why .* is fail but [^-]* success? I didn't say it was correct, I said it will work most of the time. In the "second answer", we are looking for a string match that starts with and does not have "-" in between it. So it will fail, for example, with nested comments, or comments that contain a dash for some reason, or perhaps other formats. However, it is probable that it will work for most people, most of the time. If you are one that uses "-" as a convention in comments, you will have to change the code. Regarding why .* fails, I don't know exactly. traditionally the "." character doesn't work past lines and the "*" character is "greedy" (you'll have to look that up, because this is not a regular expression forum). So, the .* should not work at all for this purpose. However, since WebDNA's grep implementation is not POSIX standard, and is a hybrid of sorts, it appeared to work in my quick test that I did before I originally posted... parsing more than one line. A further test showed that it does not work consistently. A yet further test showed that it does not appear to be "capacity limit" thing. Frankly, my real answer is neither "solution". I don't think regular expressions (in WebDNA's grep, or POSIX regex) is the "sure-fire" answer to this problem. However, one could probably build a more thorough regex call (I'd probably call the POSIX regex with [shell] via sed (find/replace) rather than use WebDNA's hybrid grep which is not yet well documented) that would be more versatile and robust, but I think the real answer probably lies in a tried and tested module/function somewhere..probably callable from WebDNA's [SHELL] or what-have-you. Though WebDNA could indeed be the solution for such a problem, *existing* modules are usually the result of many years of work by more than one person, which makes it worth checking out the third party solution. For a solution in WebDNA, I guess I'd start building a parser.. perhaps using listwords and other tools. However, it could be a long rode to get something solid. Since it is not easily done, WSC may look into this to see if something could be done with removehtml... [removehtml comments] or something. This may be worth the effort because it is something that comes up in our line of work occasionally. However, it would fall back on the priority list quite a ways. Donovan -- =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o DONOVAN D. BROOKE Pres., EUCA Design Center
WEB:> http://www.euca.us =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
Associated Messages, from the most recent to the oldest:
Owkaye Go wrote: >> (just under 800 lines) and used: >> and it successfully parsed out all the > > I think you say is correct and original answer > wrong. I will test and report my results. Can you > explain why .* is fail but [^-]* success? I didn't say it was correct, I said it will work most of the time. In the "second answer", we are looking for a string match that starts with and does not have "-" in between it. So it will fail, for example, with nested comments, or comments that contain a dash for some reason, or perhaps other formats. However, it is probable that it will work for most people, most of the time. If you are one that uses "-" as a convention in comments, you will have to change the code. Regarding why .* fails, I don't know exactly. traditionally the "." character doesn't work past lines and the "*" character is "greedy" (you'll have to look that up, because this is not a regular expression forum). So, the .* should not work at all for this purpose. However, since WebDNA's grep implementation is not POSIX standard, and is a hybrid of sorts, it appeared to work in my quick test that I did before I originally posted... parsing more than one line. A further test showed that it does not work consistently. A yet further test showed that it does not appear to be "capacity limit" thing. Frankly, my real answer is neither "solution". I don't think regular expressions (in WebDNA's grep, or POSIX regex) is the "sure-fire" answer to this problem. However, one could probably build a more thorough regex call (I'd probably call the POSIX regex with [shell] via sed (find/replace) rather than use WebDNA's hybrid grep which is not yet well documented) that would be more versatile and robust, but I think the real answer probably lies in a tried and tested module/function somewhere..probably callable from WebDNA's [shell] or what-have-you. Though WebDNA could indeed be the solution for such a problem, *existing* modules are usually the result of many years of work by more than one person, which makes it worth checking out the third party solution. For a solution in WebDNA, I guess I'd start building a parser.. perhaps using listwords and other tools. However, it could be a long rode to get something solid. Since it is not easily done, WSC may look into this to see if something could be done with removehtml... [removehtml comments] or something. This may be worth the effort because it is something that comes up in our line of work occasionally. However, it would fall back on the priority list quite a ways. Donovan -- =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o DONOVAN D. BROOKE Pres., EUCA Design Center WEB:> http://www.euca.us =o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o=o
Donovan Brooke
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
[CART] inside a [LOOP] (1997)
OT: Let's see your workspace... (2005)
EIMS Problems (1997)
subscribe (2007)
multiple databases (1997)
emailer (1997)
Sku numbers (1997)
removing programming logic white space (2003)
[LOOKUP] (1997)
Error Lob.db records error message not name (1997)
NT - Thanks and Taxes (1997)
setlineiems and UnitShip Cost (2000)
Not really WebCat (1997)
Max Record length restated as maybe bug (1997)
Webstar 1.3.1 PPC (1997)
Languages (1997)
Dreamweaver (2002)
Order not created error (1997)
Add a field to the error log? (1997)
WebCatalog NT beta 18 problem (1997)