Re: Search bug. Help
This WebDNA talk-list message is from 2002
It keeps the original formatting.
numero = 43513
interpreted = N
texte = To elaborate on Ken's group search suggestion, you could do a multi-value text search on a group that contains only CT, then specify a numeric sort on CT itself. Try this:[search db=loose-dia.db&eqshapedatarq=rd&group1field=CT&wogroup1datarq=.20, .25&group1word=ww&group1wbrk=,&CTtype=num&wocolordatarq=a,b,c&woclaritydatarq=vvs1,v vs2&colorsort=1&colorsdir=as&claritysort=2&claritysdir=as&ctsort=3&ctsddir=a s&startat=[startat]&max=[themax]]- brianAt 9:20 PM 9/17/02, marc malacarme wrote: >First I'd like to thank you for tacking the time to help. > >The reason I can't do multiple searches is because the results has about 100 >results per search and the sorting is critical. > >Here is the full search: >[search db=loose-dia.db&eqshapedatarq=rd&woCTdatarq=.20, >.25&CTword=ww&CTwbrk=,&CTtype=num&wocolordatarq=a,b,c&woclaritydatarq=vvs1,v >vs2&colorsort=1&colorsdir=as&claritysort=2&claritysdir=as&ctsort=3&ctsddir=a >s&startat=[startat]&max=[themax]] > >You see that ct is the last to be sorted. I thought about nesting searches >but it is not feasible because the of the amount of searches it would have >to perform. This is my main search page meaning that many users will be >accessing the page simultaneously and 100 searches per request would be too >much for my server. Not to mention that the page is already littered with >lookups and 3 other searches on other databases. > > I can't do a range search because I want users to be able to omit certain >values that might fit in the range. The way the user decides what values to >search Ct with is with check boxes in the form .25 .33 .50 .60 .50 .75 .100. >They can select any combination not necessarily in order. > > > > > > > > > > > >>> How else are you going to tell webcat that these are >>> two distinct values be it text or num. >> >> How about a group field search, have you tried that yet? >> Or using showif's and hideif's in your founditems context? >> >> >>> I can assume that ww, ws and wa are all for searching text. >> >> But as soon as you put &type=num into your search context, you are >> no longer searching text. >> >> I had a debate with SMSI about this issue a very long time ago. >> The HTML docs STILL do not make it clear that when you put a field >> type parameter into your search, it forces the search to behave >> differently than without it -- even though it should only be the SORT >> that is done differently ... >> >> The docs still need to be fixed to properly reflect the manner in >> which the search is messed up by the addition of a type=num >> parameter in this situation. But until the docs are fixed, it's >> always going to be the users who have to deal with this issue again >> (and again) ... >> >> >> >>> As far as running 2 consecutive search to find 2 numeric values in the same >>> field sounds a little silly. >> >> I don't think it's silly if it works! >> >> My policy when coding webdna is simple: >> >> If you cannot get it to work the way you first try it, don't waste >> too many hours trying to force it to work the way you insist that it > > should. Instead, just try a different technique -- there's usually > > another technique that works -- until the bug is fixed or until you > > can afford to invest the time to make it work the way you first >> tried. >> >> You have to remember, webdna is extremely fast, and several search >> contexts on a page will not phase the webdna search engine. In other >> words, you won't even be able to measure (much less see) the speed >> difference when searching twice vs. searching once. So if it's >> EASIER to do two searches, and if it WORKS PROPERLY when doing two >> searches, I see no reason why you shouldn't do it that way. >> >> And of course there are even more ways to skin a webcat than I >> suggested and illustrated in my previous post ... >> >> For example, you can also use showif's and/or hideif's to remove the >> ct values you don't want to include in your search results, which >> leaves you with the SINGLE SEARCH that you seem to prefer: > > >> [search db=mydb.db&neCTdatarq=[blank]&asctsort=1&cttype=num] >> [founditems] >> >> [showif [ct]=.20] >> some stuff here ... >> [/showif] >> >> [showif [ct]=.25] >> more stuff here ... >> [/showif] >> >> [/founditems] >> [/search] >> >> >> Sincerely, >> Kenneth Grome >> >> --------------------------------------------------- >> WebDNA Professional Training and Development Center >> 175 J. Llorente Street +63 (32) 255-6921 >> Cebu City, Cebu 6000 kengrome@webdna.net >> Philippines http://www.webdna.net >> --------------------------------------------------- >> >> ------------------------------------------------------------- >> This message is sent to you because you are subscribed to >> the mailing list
. >> To unsubscribe, E-mail to: >> To switch to the DIGEST mode, E-mail to >> >> Web Archive of this list is at: http://search.smithmicro.com/ > > >------------------------------------------------------------- >This message is sent to you because you are subscribed to > the mailing list . >To unsubscribe, E-mail to: >To switch to the DIGEST mode, E-mail to > >Web Archive of this list is at: http://search.smithmicro.com/ ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/
Associated Messages, from the most recent to the oldest:
To elaborate on Ken's group search suggestion, you could do a multi-value text search on a group that contains only CT, then specify a numeric sort on CT itself. Try this:[search db=loose-dia.db&eqshapedatarq=rd&group1field=CT&wogroup1datarq=.20, .25&group1word=ww&group1wbrk=,&CTtype=num&wocolordatarq=a,b,c&woclaritydatarq=vvs1,v vs2&colorsort=1&colorsdir=as&claritysort=2&claritysdir=as&ctsort=3&ctsddir=a s&startat=[startat]&max=[themax]]- brianAt 9:20 PM 9/17/02, marc malacarme wrote: >First I'd like to thank you for tacking the time to help. > >The reason I can't do multiple searches is because the results has about 100 >results per search and the sorting is critical. > >Here is the full search: >[search db=loose-dia.db&eqshapedatarq=rd&woCTdatarq=.20, >.25&CTword=ww&CTwbrk=,&CTtype=num&wocolordatarq=a,b,c&woclaritydatarq=vvs1,v >vs2&colorsort=1&colorsdir=as&claritysort=2&claritysdir=as&ctsort=3&ctsddir=a >s&startat=[startat]&max=[themax]] > >You see that ct is the last to be sorted. I thought about nesting searches >but it is not feasible because the of the amount of searches it would have >to perform. This is my main search page meaning that many users will be >accessing the page simultaneously and 100 searches per request would be too >much for my server. Not to mention that the page is already littered with >lookups and 3 other searches on other databases. > > I can't do a range search because I want users to be able to omit certain >values that might fit in the range. The way the user decides what values to >search Ct with is with check boxes in the form .25 .33 .50 .60 .50 .75 .100. >They can select any combination not necessarily in order. > > > > > > > > > > > >>> How else are you going to tell webcat that these are >>> two distinct values be it text or num. >> >> How about a group field search, have you tried that yet? >> Or using showif's and hideif's in your founditems context? >> >> >>> I can assume that ww, ws and wa are all for searching text. >> >> But as soon as you put &type=num into your search context, you are >> no longer searching text. >> >> I had a debate with SMSI about this issue a very long time ago. >> The HTML docs STILL do not make it clear that when you put a field >> type parameter into your search, it forces the search to behave >> differently than without it -- even though it should only be the SORT >> that is done differently ... >> >> The docs still need to be fixed to properly reflect the manner in >> which the search is messed up by the addition of a type=num >> parameter in this situation. But until the docs are fixed, it's >> always going to be the users who have to deal with this issue again >> (and again) ... >> >> >> >>> As far as running 2 consecutive search to find 2 numeric values in the same >>> field sounds a little silly. >> >> I don't think it's silly if it works! >> >> My policy when coding webdna is simple: >> >> If you cannot get it to work the way you first try it, don't waste >> too many hours trying to force it to work the way you insist that it > > should. Instead, just try a different technique -- there's usually > > another technique that works -- until the bug is fixed or until you > > can afford to invest the time to make it work the way you first >> tried. >> >> You have to remember, webdna is extremely fast, and several search >> contexts on a page will not phase the webdna search engine. In other >> words, you won't even be able to measure (much less see) the speed >> difference when searching twice vs. searching once. So if it's >> EASIER to do two searches, and if it WORKS PROPERLY when doing two >> searches, I see no reason why you shouldn't do it that way. >> >> And of course there are even more ways to skin a webcat than I >> suggested and illustrated in my previous post ... >> >> For example, you can also use showif's and/or hideif's to remove the >> ct values you don't want to include in your search results, which >> leaves you with the SINGLE SEARCH that you seem to prefer: > > >> [search db=mydb.db&neCTdatarq=[blank]&asctsort=1&cttype=num] >> [founditems] >> >> [showif [ct]=.20] >> some stuff here ... >> [/showif] >> >> [showif [ct]=.25] >> more stuff here ... >> [/showif] >> >> [/founditems] >> [/search] >> >> >> Sincerely, >> Kenneth Grome >> >> --------------------------------------------------- >> WebDNA Professional Training and Development Center >> 175 J. Llorente Street +63 (32) 255-6921 >> Cebu City, Cebu 6000 kengrome@webdna.net >> Philippines http://www.webdna.net >> --------------------------------------------------- >> >> ------------------------------------------------------------- >> This message is sent to you because you are subscribed to >> the mailing list . >> To unsubscribe, E-mail to: >> To switch to the DIGEST mode, E-mail to >> >> Web Archive of this list is at: http://search.smithmicro.com/ > > >------------------------------------------------------------- >This message is sent to you because you are subscribed to > the mailing list . >To unsubscribe, E-mail to: >To switch to the DIGEST mode, E-mail to > >Web Archive of this list is at: http://search.smithmicro.com/ ------------------------------------------------------------- This message is sent to you because you are subscribed to the mailing list . To unsubscribe, E-mail to: To switch to the DIGEST mode, E-mail to Web Archive of this list is at: http://search.smithmicro.com/
Brian Fries
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:
Another new version? (2000)
Email header munger in [sendmail] (2004)
Retrospect Backup Misses Empty Folders (2000)
[OT] - See my site? (2004)
form data submission gets truncated (1997)
Help with Shipping Costs (1997)
4.51 Triggers (2003)
Bug Report, maybe (1997)
[WebDNA] Fails to recognize comma-separated db (2010)
WebCommerce: Folder organization ? (1997)
ConvertChars - german umlauts (1998)
Colors (2000)
WebCat2b13MacPlugin - nested [xxx] contexts (1997)
About the hyphen in searches (2003)
Format of Required fields error message (1997)
Before I Can Begin . . . (1998)
WebCat2 - [SendNews] (1997)
(2000)
Updating Prices in DB (working) (1999)
Help! WebCat2 bug (1997)