and....one of the biggest reasons that i became a sponsor of the kde port...is that im regex dumb, and am a visual learner... ill learn regEx quite soon after i see more of these "translations"
its the translations, that turn whats the definition of "blah" into "english" is what teaches me more than anything. and surprisingly either im not looking or there are none out there...that is good tutorials on regEx. i mean, it seems like its a very obscure science in and of itself, and therefore like that of a cult status, and regEx competitions its not too prevalent. maybe im wrong, thats just my take. but it will be neat to see how this tool evolves, and im stoked i was able to help fund the first iteration, for what could turn into something rather large, may web servic-able? might we all see a wsdl in this ones future...i hope so...and this group could do it... a webservice based regEx and remoting tool, that prompted the user with a series of questions it "knows" to ask based on previous assertions? anyway, just an idea, and if it flys, i penned it well, at least first today ;) anyway...thats just why i posed the question/suggestion. im <cfset tony = reFind(f***inDumbWhenItComesToregEx)> adios tw -----Original Message----- From: Michael Dinowitz [mailto:[EMAIL PROTECTED] Sent: Sunday, July 13, 2003 9:11 PM To: CF-Talk Subject: Re: (OT) RegEx editor I'm typing this from my laptop as my computer has died. This has 2 main effects. The first is that I don't have a good list of the sponsors of the RegEx editor on me and the second is that I don't have the source code for it. On the other hand, I'll be able to get their names off of paypal and the windows executable is on a server. Bottom line is that tomorrow you'll be able to both see who has supported the creation of the visual RegEx editor (and sponsor it yourself if you want) as well as download it for yourself. Once my computer is up again I can work on getting a Mac and Linux version of it (seperate from the full KDE). That being said, lets look at Tony's request. What's needed is an interigator that will build the RegEx based on the answers to some questions. Lets take an example and the questions asked from it: [EMAIL PROTECTED] 1. Is this the start of the line or can it be anywhere in a string? 2. Does the text before the @ sign (chosen as being non-text) have to be only those characters or can they be any characters? 2a can the text be numbers as well? 2b can the text contain any non-letter/number characters? 2c is there a specific length to this segment? 2d must it start with a specific type of character? 2e must it end with a specific type of character? Here are the answers: 2a - can be numbers 2b - can contain the characters -_. 2c - must be at least 1 character long 2d - must start with a letter or number (note may conflict with 2c and have to be careful) 2e - no Just based on that we'd have this before the @ sign [a-z0-9][a-z0-9_.-]* so far so good. It's a legal start of an email address (not a full one with a text name as well, just the address). Now we look after the @ sign and we see another non-text character (a period). The next questions will have to deal with that and will follow the same mold. The only additional questions are: 2f can this segment be repeated? (yes) 2g must each segment end in a period? (yes) This will give us a total of: ([a-z0-9][a-z0-9_-]\.)+ The final section will have the following answers: 2a only letters 2b no 2c yes 2f no and extra questions of 2h is there a set length to this segment? (yes) 2i is there a minimum (yes, 2) 2j is there a maximum (yes, 6) and the end result will be [a-z0-9][a-z0-9_.-]*@([a-z0-9][a-z0-9_-]\.)+[a-z]{2,6} This is a perfect regex for an email address and it can be built with questions alone. Some of the more advanced stuff might be a lot harder to write using questions, but it can be done. > Just as a point of reference, the actual regex for your whatever is > [a-z]{4}[0-9]{4}[a-z]{4}[0-9] > > However, the real question is, how would the system know (for example) that > you meant for X and 1 to be indicative of [a-z] and [0-9] instead of a > literal capital x and the literal numeral 1? Language theory states that > any system you create in this way must have a syntax as complex as (or more > complex than) the maximum complexity of regular expression it can create. > About the best you can do is make the system easier to use by doing things > like making the syntax more user-friendly or creating a visual editor for > them. Oh, wait. Michael's got that second one in progress. :-) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~| Archives: http://www.houseoffusion.com/cf_lists/index.cfm?forumid=4 Subscription: http://www.houseoffusion.com/cf_lists/index.cfm?method=subscribe&forumid=4 FAQ: http://www.thenetprofits.co.uk/coldfusion/faq Get the mailserver that powers this list at http://www.coolfusion.com Unsubscribe: http://www.houseoffusion.com/cf_lists/unsubscribe.cfm?user=89.70.4
