You probably don't mean to say this: eventDate >= #CreateODBCDate("07/12/2005")# AND eventDate = #Days# Undoubtedly you intended to use a different variable in the second case. ... View more

I do not see how the RewriteRule that you have specified here would accomplish the substitution that you are describing. Here's an online tester that you can use to work out the problems in your rewrite rules: http://civilolydnad.se/projects/rewriterule/ Saves a lot of time. ... View more

I suggest that you look closely at the MySQL function library reference material for the version of MySQL that you are running ... and, consider upgrading if you need to. In any case, you might be able to get more precision by asking for it, e.g. with the CAST() and CONVERT() functions. The MySQL documentation (as recently as 5.0) discusses precision math functions. Most significantly, though, it seems that MySQL now offers a library of geo-spatial (GIS) functions which might make it possible for you to go well beyond the Pythagorean Theorem in addressing your problem.  The boundaries of each ZIP Code are, of course, a closed polygon, and what you are really looking for here is to determine containment of a chosen point within this and/or adjacent polygons.  MySQL 5.0, for instance, appears to have considerable direct support for this.  You do have to purchase these data sets. See:  http://dev.mysql.com/doc/refman/5.0/en/spatial-extensions.html ... View more

When you define any Windows service, one of the "properties" of that service is the definition of the account that it is to log-on as. (See:  Administrative Tools .. Services.  Choose a service, right click, choose Properties.  From the pop-up, choose the Log On tab.  Your Windows version may vary.) By default, services tend to log on as the "Local System account," which probably isn't what you want, I think, for something like CF.  Not if it has the ability to interact with local files, e.g. uploading and downloading.  It is far better in my opinion for you to define a local account ... one that is not capable of being "logged on" to ... whose only purpose is for it to be used by the CF Server.  This will allow you to build a definite boundary around what the CF Server can do, and it will also let you easily identify the files that are created by the server.  The so-called "principle of least privilege" is a good bright-line rule to follow, and when CF is being used to manipulate files (locally and/or elsewhere), I think it's prudent to give CF its own explicit identity. ... View more

This problem is necessarily caused by modification of the server-side script files.  Therefore, it is necessary that the server must have been compromised.  On a shared server, this is "more or less to be expected," since hundreds if not thousands of people other than yourself have accounts on the same server(s).  However, there is a lot that you can do to preven it:  it only takes a couple of extra steps. Always use sftp (secure FTP) to transfer files and to establish host sessions.  Many hosting companies allow you to disable non-secure FTP, and you should do so. Protect your username/password as robustly as you can.  "password1234" is an extremely common password, as is "password."  After transferring files to your computer, use chown (Unix), cacls (Windows), or its equivalent to secure the files against anyone, including yourself.  (You can always countermand your directives later, the next time you're ready to update the site.)  This will at-least require the intruder to have compromised your account, not someone else's. In that same vein, don't forget to secure all of the parent directories!  "It'll do no damn good at all" if the intruder can replace your carefully-secured directory, in its entirety, with an altered one! Make backups! If your site permits any sort of user-contributed content to be provided, make absolutely sure it is well-filtered. "Think like a pirate."  ("RRRRRrrrrrrrr!!")  Web-site compromises are always "crimes of opportunity."  Many a cat-burglar has made a good living by trying each door in a neighborhood while carrying a pizza box. ... View more

(Shrug...)  I used DW as an HTML/ColdFusion editor and never employed any other part of that product.  Eclipse, by the way, is absolutely free and does just as well.  I just don't grok the notion that you can really expect "whizzy-whig" when there's code involved in the scenario. In short:  it don't surprise me that you're seeing this, and I counsel you to avoid the "fee-chur" that's leading to it. ... View more

Please don't expect this forum to teach you how HTML form-submission works.  Whether you do it by "Googling it" or by reading tea-leaves, you're going to have to put forth a little bit of mental effort here.  There are ga-zillions of tutorials on this subject, and the only way that you're going to get a handle on what is going on is to study a few of them.  Everything that you'll read in any tutorial will apply, "ColdFusion or not." If I were teaching a class tonight, I would put this on my pop-quiz:  (note to my students: "at ease, no pop-quiz this week...") 14.     In the space below, using no more than six sentences, explain what happens when the following scenario occurs:  (a) a web-site displays an input form, then (b) the user fills out that form and presses a "Submit" button, then (c) the web-site processes the input and sends back a response.  (The "scenario" you are to describe consists of steps (a) through (c) taken together.) I mean it... in your own words, what t'hell actually happens?  You've got two computers here (the user's, and the web-server), and they are talking to each other.  What do they actually say to each other at each step?  Go find that answer.  (Elsewhere.)  Understand it thoroughly.  Then, come back and reconsider your ColdFusion-specific question.  "The little light-bulb will come on, and when it does, it will stay on." ... View more

Consider how HTML works.  If we leave JavaScript out of the picture for simplicity, the user types his information into an input-box and then presses some kind of a submit-button.  Both the input-box and the button must therefore be contained in a <cfform>.  The web browser will package-up the entered data into (probably) POST data and send it to whatever URL is designated as the target of the form.  This target, then, will be the same or a different ColdFusion .cfm page. This page, therefore, is "where you begin."  Whatever data the user has entered will be found as an appropriately-named variable in the POST data, along with the identity of the button that he just pushed.  This form must collect the data, validate it, and if it is valid send it to Twitter.  (You will use the same tag that you described, but the text comes from a POST variable.)  Then, it must either display some kind of HTML output, or use the <cflocation> tag to redirect to some other page which will do so. ... View more

In most cases, you do know what struct-keys to expect.  When you don't, you can usually get what you want by <cfloop>ing over the structure itself.  You can also, of course, obtain a list of the keys. To handle the case of bumping into a struct-key that doesn't exist, well, you've got two good choices ... depending on how probable you consider the event to be, and/or if the presence or absence of a key "means something." The key-exists function will tell you if "the bridge is out" so that you don't wander over a precipice.  The "tradeoff," theoretically at least, is that you are probing the structure for key-existence twice in a row.  Now, hash-lookups are extremely fast by their nature, and the CF/Java implementation is known to be "good." Exception-catching via <cftry> is another viable choice.  I happen to prefer to put such a block around everything, so that my app doesn't "lose face" if when something happens to go wrong, but exceptions do have a place in ordinary procedural logic.  You simply place a nice comfortable trampoline under the bridge, then charge right across it.  Most of the time the bridge holds-up just fine and you haven't wasted any time checking the floorboards.  When something "exceptional" does occur, however, you fall through the deck, land harmlessly on the trampoline and "respond appropriately."  If you find yourself in the exception-handler, you know that whatever you were looking for just happened. It all pretty-much coms down to your personal preference, because ColdFusion is typically not used to handle computationally-intensive nor time-consuming work.  Whatever makes your code cleanest and most-reliable is the path that you should take. ... View more

You might have to tweak the result by one depending on whether today's date is before or after the birthdate... ... View more

You need to be using the <cfqueryparam> tag. Do not construct SQL queries by the blind and naiive concatenation of string values, especially not values obtained from the user himself. This tag will present the value as a parameter to the SQL query, magically providing the query-string with a "placeholder" in the right location and then separately providing the value corresponding to that placeholder.  In this way, the values presented by the user can never be misinterpreted as part of the SQL command itself. ... View more

Oops.  Sorry.  Wrong thread. ... View more

It might be... but what if instead you catch the error (if it occurs) and do something to deal with the error on the rare occasion when it happens? From your description, it seems that the problem is "the connection has been lost."  Okay, that's a very-occasional thing (so you say...), doesn't happen too often, and so the real issue (one might say...) is that the end-user sees it happening with a very-ugly error message. If you frame the logic in a try..catch block, the exception will be trapped when it occurs and then you can take whatever recovery-action you might need to do (e.g. re-establish the connection and try again...).  Most of the time, the exception doesn't happen and you just go on your merry way, having not wasted any time laboriously checking to see whether the error happened or not.  You just stick someone out there in right-field to catch the ball if, this time, it happens to "go foul."  It costs next-to-nothing to post him out there, and he only has something to do when-and-if "the rare circumstance that he is looking for" actually happens.  (You don't have to say ... "if the ball went foul then..."  If the ball went foul, as it does 2%-or-less of the time, then your right-fielder unerringly catches it and that's how you know.) Offhand, I think that this would be a lot less costly than maintaining a bunch of connections.  (I opine that such an approach would prove to be a decision that you regret for other reasons.) The term, "exception," is well-named.  Take advantage of them. ... View more

As usual, there is no "one true answer."  It's a trade-off, and it comes down to this: On the one hand, CFCs remove code from "the web page" (CFM), where it might be taking up a lot of space, and put it into a separate location where you can focus upon it when you want to, and completely ignore it when you want to. On the other, CFCs create "coupling" between different parts of the site which may or may not, given the situation, be a good thing. Given that there is no "bright-line rule" to be found here, my best advice is simply:  "whatever you do, plan what you do, and maintain that plan throughout the service-life of the application."  (Most likely, the application's tenure at the company will considerably out-live your own tenure at the same company, so your plan serves many distinct purposes.)  Contemplate your next move before you make it. ... View more

The only way that hackers can modify your ColdFusion Code is to get access to your server so that they can actually modify the CFM and/or CFC files that are stored upon it.  Unfortunately, on a shared-hosting setup that's not terribly difficult to do.  And, most programmers neglect to consider the file access permissions (other than the "x"ecutable bit in Unix/Linux) that they attach to any particular file when they upload them. You need to be certain that all of the files in your directories, and the directories themselves, are locked-down so that no one can modify them, and so that no one but "you 'n the web server" can see what they contain.  (Remember, shared-hosting companies give away shared-hosting accounts like water, and it's sometimes effortless for "the web-site next door" to see much more than it ought to be able to see... and maybe, to modify something!) If you're running on a Linux host, see if the server appears to support Access Control Lists (ACLs) and whether you as a secure-shell user can establish them.  If so, this will allow you to restrict access more thoroughly than the "owner/group/world" permissions-mask system would allow.  The equivalent mantra with regards to Windows hosts is different in details only.  One way or the other, implement the "principle of least privilege." The larger problem, of course, remains with us:  the end-user's computer, and the shameless reality that the aforesaid user is probably an all-powerful Administrator of a Windows "Home Edition" something-or-other ... whose entire system, therefore, is a sitting duck with no backups.   You can't do anything at all about that. ... View more

You might be able to approximate this with some JavaScript ... but offhand I would suggest that you re-think your approach. ... View more

MySQL provides a non-standard REPLACE INTO syntax which glosses over this issue. However... I prefer to use stored procedures and views for all database activities these days, mainly because it puts all of the database-related logic and dependencies into the database itself... where I can always find them and can always know what is actually going on.  Put the necessary logic, whatever it is, into the stored procedure. ... View more

It is usually the case that when the user opens "another browser window or tab," all of these are owned by a single instance of the web-browser, present the same set of "cookies," and therefore effectively share the same state.  (Which is nice for me right now because I did not have to log on to the forum in every Firefox window that's looking at this forum right now.) This is pure-and-simply an issue of "how HTML works" vis-a-vis "your application's design." ... View more

ColdFusion isn't Perl, but its structs are, effectively, "a hash."  Therefore, it seems to me that instead of "looking for," say, "27," you should use a hash struct to determine if the value is there.  Obviously, you can do both:  the data value in the hash struct could be the [X,Y] dimension in the array where that particular value can be found. AFAIK, structs can be multi-dimensional so that (Holy REXX, Batman!) you could find yourself querying, say, mystructure.27.X and mystructure.27.Y... ... View more

If this is a standard HTML form, then it necessarily is going to submit once, and the value submitted (hence, seen by CF) will be "other."  What you choose to do with that is up to you. One thing that occurs to me is that you could add a little custom JavaScript to the form that would specifically look for "other" and pop-up a client side input box to gather the value.  The JavaScript would then install the value into a hidden submit-field.  Thus, when CF gets the posted record, it will see "other" and the Javascript-supplied choice. There might be a funky way to do this automagically ... CF has many JavaScript tricks up its sleeve ... but I do not know what it is.  "Bound fields?"  I dunno.  Either way the special-effect is going to be achieved using JavaScript:  the question becomes whether you have to be the one to write it. ... View more

It appears to me, from http://livedocs.adobe.com/coldfusion/8/htmldocs/help.html?content=sharedVars_13.html that the most appropriate strategy would be to use Session.SessionID.  It seems to me, based upon this reading, that "the designer's intent" was that this would be a unique-identifier string that will reliably work in all cases, i.e. whether Session or J2EE persistence methods are used. ... View more

Funny thing about technologies:  once they get adopted by a company, they're deployed hundreds or thousands of times ... and what's more, they work! X years later, you find that they are still working.  This is especially true of timesharing web-deployed systems, where you really only have to get the thing running properly on the big iron blade server(s). The reality of the computing world hasn't changed much in fifty years: Once you've got something working properly (especially when it runs on just onecomputer, namely a mainframe server), it will probably continue running much longer than, well, than you will.    ("Dig me up and I'll fix it then...") Change, of any sort, is uncertain, disruptive, and expensive.  In any case, "change for change's sake" has no ROI. So, companies acquire quite a "hodge-podge" of systems, review them carefully, tend them carefully, and more-or-less-willingly use them quite successfully for years. The server computers still sit in enclosed rooms with lots of air-conditioning and raised floors.  (And now-and-then are programmed in COBOL.) So...  when you walk into any new "shop," you take a look around and "do as the Romans do."  Grab a manual on Friday and be competent by Monday. People tend to say that their companies are only using the neatest, coolest stuff when they're talking on the phone to journalists.  (And I happen to think that ColdFusion does qualify as "neat, cool stuff!"  For what it's designed to do ... which is a very commonly-occurring set of requirements ... it actually rocks.) ... View more

The answer to your quest is found in the very words that you use:  what does "the last record" actually mean to you? Conceptually, the rows in an SQL database have no inherent "order."  There is no express concept of a newly-inserted record being "anywhere in particular," nor does whatever physical record-position that the records may possess actually "mean anything."  (This was the dramatic improvement that SQL databases made over their linked-list predecessors such as IMAGE and IDMS.) Records are frequently filed under a primary-key that is an ascending integer, but this property should merely be regarded as an artifact, subject to change without notice.  A primary-key value contains no inherent information:  its numeric (or what-have-you) "value" means nothing. When you say "the last record," I think that you are saying, "the most recent one."  That is to say, the one(s) with the largest value for transaction-date.  You can accomplish this search with a subquery, which asks for the "MAX(transaction_date) GROUP BY customer_id" or what-have-you.  The record(s) that you are looking for, then, are those whose (transaction_date, customer_id) are "IN" that set of records. When SQL is presented with a query, of arbitrary complexity (including subqueries), it always evaluates the entire request and figures out the optimal way to accomplish it ... which it will tell you about in exhaustive detail by means of the EXPLAIN verb. A very nice recent treatment of this task can be found at: http://www.xaprb.com/blog/2006/12/07/how-to-select-the-firstleastmax-row-per-group-in-sql/ ... View more

Interesting.  It is inconsistent with my documentation that "Yes" (a boolean) would be an acceptable value for this parameter.  My docs say it's supposed to be an integer. The server-logs (e.g. Windows event-viewer) ought to be throwing more-detailed error records when CF attempts to do the LDAP connection and fails to do so. Referrals are pretty-much a fact of reality in LDAP configurations in the real business world, where equipment moves around a lot and network topology changes, but they do take extra time.  If you can point your applications directly to the correct final-location of where the desired objects are actually now stored, that would be an improvement. ... View more

"Holy COBOL, Batman!" The way that we did such things, in the good ol' COBOL days (actually, since the days of Herman Hollerith...) was to sort the data by each of the major keys and then detect when the values changed. For instance, if your query includes ORDER BY STATE, your_remaining_keys, it will now automagically produce output that is in the order you're now accustomed to, for each STATE.  Therefore, all of the rows for Alabama will be together, followed by all of the ones for Alaska, then all of the ones for Arizona, and so-on all the way to Wyoming, no matter how many STATEs there actually turn out to be.  Furthermore, once you have seen the last of Alabama, you know that you will never see anything from that state again. Therefore, add logic in your loop that detects when the STATE changes from the previous record.  Something like this... <cfloop ...>    <cfif NOT IsDefined("previousState")>    <!--- this is the case the first time through --->       <tr>       ... any other HTML output you need here ...       <cfset previousState = looprecord.State>    <cfelseif looprecord.State IS NOT previousState>       </tr>       <tr>       <cfset previousState = looprecord.State>    </cfif>    ... </cfloop> <cfif IsDefined("previousState") >    <!--- loop ran at least once>    </tr> <cfelse>     <!--- any handling you need for a completely-empty recordset ---> </cfif> Since we (must!) know that the records are sorted by State, this logic detects when the State in the current record is different from the one in the previous record that has been encountered.  Note the following: The first IF-case handles the very first record in the loop, when there is no previous record (and also, it has been established that the query is not completely empty).  This is the start of the first group of records, and the start of the group of records for this state. The second IF-case handles a transition from one State value to the next.  This is both the end of the group of records represented by previousState, and the start of the group of records for the next one. The last CFIF, after the loop, handles the case where we have reached the end of the records and, by virtue of the fact that "previousState" has some value, we know that the query was not completely empty. This is the way that massive amounts of data are routinely handled in the "COBOL days" when a computer was a magical machine in a huge room in science-fiction movies.  It is also the way that Herman Hollerith devised to produce US Census results long before computers were invented. This strategy will allow you to handle an unlimited number of states with no need for separate queries for each one. If you need to handle "all 50 states, present or not," simply do a LEFT OUTER JOIN of your data against a table that has all possible state-values in it.  (Make very sure that all possible states do exist in this table.)  You can now test for transitions of this state-name column ... and by looking for NULL values in appropriate columns you can know whether it was joined to real data ("this state is present in the data") or not ("this state is missing"). A non-SQL version of the same idea would rely upon a ColdFusion array containing all of the known state-names in ascending order:  the array can be used to recognize and fill-in any gaps, during the output process previously described, eliminating the need for a join.  (It can also be used to detect the occurrence of a state-name that is not valid.) ... View more

What are the sort of "errors that might suggest that sessions are getting lost?" It's very helpful to include a customer-feedback "contact the webmaster" link ... one that is as technically-uncomplicated as possible ... because when you do so, customers will often tell you about what just happened to them; what went wrong.  Although you get a fair amount of fluff and spam from such feeds also, it can alert you to problems that you don't know about (and greatly help to diagnose the ones that you do). ... View more

An application in "simple ASP" should be straightforward to convert to CF, but it will nevertheless be a manual process.  There won't be a one-to-one correspondence between the ASP code and the CF code:  the CF equivalent is likely to be much smaller and simpler. The difference (and the potential advantage of CF) lies in its fundamental approach to the problem: ASP is, well, Visual Basic.  It's a programming language by nature, and a web-page is therefore "a script."  The web-server hands the car-keys to the page (full of executable code...) and the page drives the car to its destination. CF is, in general, much more of a declarative language.  The various documents contain a well-defined structure ... the structure of HTML itself ... and they are parsed and processed (conceptually speaking...) by an external processor which carries out the desired action based on the overall structure of the page that it sees.  Since "most pages of a certain type are more-or-less exactly the same, and there might be thousands of them," CF gives you a few well-chosen ways to build "pages like that" and to do so quickly.  CF "already knows how to drive the car to the destination," and the stuff that you write consists mostly of details. (I recognize that my description, above, of CF's actual modus operandi is not technically correct.  But the big-picture conceptual idea stands, and that's all I'm driving at here.) The best way to take advantage of a tool like CF in a conversion project is, "when in Rome, do what the Romans do."  Study CF to see what it can do "quickly and easily," and adapt your project's design to shamelessly take advantage of that way.  Do not spend a lot of time trying to figure out how to "jimmy" CF so that it will do a certain thing in a certain way. The choices that were made by CF's designers are pretty-good ones.  Although CF gives you "a full programming language if you want to use it," its greatest strength is in what you don't have to write in order to get "a certain thing" done.  If you find yourself writing a lot of programming, then you need to step-back and have another look:  "there's more than one way to do it."  CF's interpretation of a dot-NET application will be quite different from the original, and it should be.  If it isn't, you're not getting market-value out of the conversion effort. ... View more

I wonder if your AS/400 can sue you for damages, for giving it a query like that. Seriously, friend:  learn SQL.  You need to clearly understand how a query can be built and what the computer actually does in order to carry it out.  This query is grotesque. (1) When you list multiple tables in FROM without a JOIN clause, you risk a "cartesian product."  (Google it.) (2) You have no need for repeated sub-SELECTs when you could not only calculate all the aggregate values in one sub-SELECT, but could actually eliminate the use of sub-SELECTs altogether. (3) You need to clearly understand how your query translates into the 'execution plan' that the query engine actually uses.  You need to seriously learn how to be your computer's best friend... it's best "query optimizer." I have seen "hour long" queries be reduced to "a handful of seconds." ... View more

Ummm...  if 500,000 Acrobat installations were falling dead with the same issue, then it might be Adobe's problem to "fix."  Otherwise, it is yours, and the question for you becomes, "how?" The first place that I would look is in the Event Viewer, which is one of the Administrative Tools and which might require Administrator-level access.  When a Windows call fails in some unexpected way (such as one that might cause a message as cryptic as this one), Windows probably makes an entry in the error-log. It would also be useful to know if any other applications have become similarly unstable at about the same time. I do not generally find it useful to blame such things on the usual knee-jerk assumption, "it must be a virus."  Taking for granted that of course you are not running your Windows box as an Administrator ... ... that really won't be much of a likely possibility.  There is something that is causing a LoadLibrary() API call to fail, under conditions that ought never to occur.  The Event Viewer is your most likely source of useful, factual information about what might be going on in your system(s). ... View more