Actually a tool like this would not be hard to write, the only downside is that it would take a very long time to run in CFML though.
It would be a very good candidate as an extension to CFBuilder, the images would be just find all with known image extensions and search for known links in the files. Not found then flag it as such. Methods would be a matter of just getting the metadata for the CFC and then searching for that method being used. Not sure how long it would take on the size of a large application, like Dale is thinking of running it on, but it would be down to the amount of files and methods that would have to be recursed on each file. -- Regards, Andrew Scott WebSite: http://www.andyscott.id.au/ Google+: http://plus.google.com/108193156965451149543 On Thu, Dec 22, 2011 at 2:50 AM, charlie arehart <charlie_li...@carehart.org > wrote: > Besides the ideas shared so far, I have another couple of thoughts. That > said, neither is the perfect answer for you, nor is there one to be honest. > There’s been talk of such “code coverage” tools for CF, but to date none > exist (as far as I know). > > As for what CFM and assets are accessed, you could determine that somewhat > using web server logs. Of course, they don’t offer the absolute path, only > a web-relative one, and if you have multiple sites that could make things > harder. Of course, that won’t help you see CFCs called (from within CFML, > though you would see CFCs called remotely, such as via web services or from > ajax, etc.) > > You’d need to do such log analysis over a fairly long time period to > increase the confidence of your resulting conclusion. > > You could also get an approximation of calls to CFM and CFCs (including > internally called CFCs) by way of the cfclasses directory (within the > [CF]\wwwroot\WEB-INF\ directory). If you have the CF Admin set to “save > class files”, then whenever a CFM or CFC is called, a class file is created > for it (in the case of CFCs, a class is created for every method within the > CFC—not every method called, but every method that exists within it, which > will not help with your point 3 of course). > > As for the accuracy of the data that would show, you could base your > decision (on what to consider “has been used”) either by what you see in > that directory today (which may reflect months or years of calls), or you > may want to clear it out and run for some period of time to see a more > recent reflection of what’s called.**** > > ** ** > > /charlie**** > > ** > -- You received this message because you are subscribed to the Google Groups "cfaussie" group. To post to this group, send email to cfaussie@googlegroups.com. To unsubscribe from this group, send email to cfaussie+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/cfaussie?hl=en.
