01.26.11
Posted in Updates at 1:24 pm by Todd Wilson
This one takes care of a couple of bugs that slipped through in the last version:
- Restored the horizontal scroll bar in the last response tab.
- Fixed an error that caused screen-scraper to disallow testing extractor patterns.
Permalink
01.25.11
Posted in Updates at 11:17 am by Todd Wilson
This one contains several fixes and enhancements:
- Fixed a bug related to hitting the “Enter” key in the find dialog box.
- You can now wrap text in the last request and last response panels.
- Rearranged elements on the last response panel so that overlapping shouldn’t occur.
- The delay on the script auto-complete box can now be set via the “AutoCompleteDelay” property in the “screen-scraper.properties” file.
- Rearranged elements in the proxy “Progress panel” so that they don’t overlap.
- Now dismissing the splash screen before the start page loads.
- The name text box is now highlighted when proxy sessions, scraping sessions, and scripts are created.
- Adjusted a few visual elements related to proxy sessions so that they resize correctly.
- Now filtering out “sitecheck” requests made by Opera.
- Table columns in the “HTTP Transactions” table are now being sized correctly even when the table is empty.
- Fixed a bug where less-than symbols weren’t always showing up in the tool-tip for extractor pattern tokens.
As always, full history and details can be found in the
Alpha Change Log. Also, for anyone keeping track, we’re getting very close to releasing another public version of screen-scraper (we’ll probably give it a version number of 5.5). If you’re running the alpha versions we’d be grateful for any
bug reports. We’ll obviously want to work out any kinks before we release the next stable version.
Permalink
01.21.11
Posted in Miscellaneous at 7:20 pm by Todd Wilson
We recently added experimental support to screen-scraper for client/PKI certificates. Some web sites require that you supply a client certificate, that you would have previously been given, in order to access them. I say this new feature is “experimental” because we’ve only been able to perform limited testing with it. So far, it does seem to be working as it should, though.
In order to account for sites that use client/PKI certificates, we’ve added a feature to screen-scraper that allows it to use JKS files. These are files used by Java that encapsulate secure certificates and such. The trick is to turn your existing client certificate file(s) into a .jks file. We’ve currently only tested the feature using .pfx files, which we converted into a .jks file via the method described here:
http://www.cb1inc.com/2007/04/30/converting-pfx-certificates-to-java-keystores/
In the current alpha version of screen-scraper, if you look under the “Advanced” tab for a scraping session you’ll see a box where you can enter the location of your .jks file, and a box that will take the password you used when generating the .jks file. There are also corresponding boxes under the “Advanced” tab for a proxy session.
If you’d like to use this new feature you’ll likely need to do some of your own research on how to turn your client certificate file(s) into a .jks file. Here are a few sites that may help you in this:
You might also find these tools to be helpful:
Because this is a particularly difficult feature to test, we’d appreciate any feedback from those who are able to try it. If you get the chance to give it a whirl, please also take a minute to
post to our forum or
send us a message.
Permalink
01.17.11
Posted in Updates at 6:17 pm by Todd Wilson
A few bug fixes in this one:
- Fixed a bug related to the anonymization service.
- The DataManager now handles reserved words correctly.
- Fixed a bug related to data extraction timeout.
- Fixed a bug related to requests being recorded with redirects.
See the
Alpha Log for the full history of changes.
Permalink
01.11.11
Posted in Updates at 1:07 pm by Todd Wilson
This one contains several changes:
- Including a UseGlobalExternalProxyForAllScrapingSessions property in the screen-scraper.properties file will now cause global proxy settings to apply to all scraping sessions.
- Fixed a minor bug related to invalid extractor pattern token names.
- Undo in certain text boxes can now be triggered properly via keyboard shortcut on a Mac.
- Now notifying the user if there are no matches when the “Highlight Extracted Data” button is pressed.
- The “Last Response” tab can now be displayed in a separate window.
Permalink
01.07.11
Posted in Miscellaneous, Updates at 3:49 pm by Todd Wilson
A few months ago we announced a drop in the price of our anonymization service for those using the latest alpha versions of screen-scraper. Unfortunately, things didn’t work out with this quite as we had planned, so, for the time being, we’re returning to our previous price of 25 cents per server per hour.
For a bit more detail, we were able to do this because Amazon announced the availability of even smaller virtual machine instances that we could use as proxies. As we’ve used these smaller instances over the past few months however, we’ve found them to be so unreliable that they’re essentially unusable. Our hope is that Amazon will improve things on their end, and, once they do, we’ll start using them again and drop the price back down. Cross your fingers and keep posted for when that day comes.
Permalink
01.03.11
Posted in Updates at 6:40 pm by Todd Wilson
Just fixed a minor bug related to cookie handling that resulted from the recent upgrade to the proxy.
Permalink
