Tuesday September 29, 2009
Version 2.1.3.2
Adobe Acrobat version 9.0 update generates PDF files that crash older ReportServer versions. We updated PDF to TXT translation module. This version also requires an update to ghostscript 8.64 for updated PDF to TIFF translation.
An issue was fixed where the installation serial number was not able to be pasted back into the ReportServer when registering a new installation.
Tuesday December 16, 2008
Version 2.1.2.5
FIRSTPAGE rule operator was added to text parser rule interpreter. This rule works like the previously added LASTPAGE operator. When FIRSTPAGE appears at the beginning of a rule, then that rule is only fired on the first page of the document.
An issue was resolved where a field on a single page document could populated two times. The first time is when the FIRSTPAGE operator is in effect. The second time is when the LASTPAGE operator is in effect. The first rule of any type that populates a field with a non-empty value is now the winner.
If the channel settings do not specify a target folder, either a done folder, or an error folder, then the channel is not halted with a fatal error. This behavior is desired for demonstration and other setups, where the post processing routing does not need to be setup. An information message is written into the channel log that states that no target folders were specified, and therefore the input file has been deleted and the channel continues with its processing.
Thursday December 4, 2008
Version 2.1.2.4
LASTPAGE operator was added to text parser rule interpreter. When LASTPAGE appears at the beginning of a rule, then that rule is only fired on the last page of the document.
A double click on a channel window now brings up a dialog that shows the text in an edit field, so that the text can be selected and copied and pasted.
Monday November 24, 2008
Version 2.1.2.3 - New Text Parse Rule
LOOKFORGETBELOW 'string' width, finds a line that contains 'string', then extracts width characters centered around the found location in the subsequent line. This rule is useful for extracting data from labeled values, where the value appears below the label.
Thursday October 28, 2008
Version 2.1.2.2 - Text Parser Can Get Values From Windows Captions
A new text parse command, FILESUMMARY has been added. This provides the ReportServer with the ability to obtain index data values from the windows file summary tab, such as author, subject, keywords, etc.
Thursday October 23, 2008
Version 2.1.2.1 - built in SAP scan algorithm
Added new IDOC command for SAP called "IDOC-RESOLVE:". This new rule provides the XML node scanning mechanism that is needed to process SAP XML IDOC documents. *Note: valid XML would not need this provision.
Monday September 15, 2008
Version 2.1.1.9 - Two new channel types
Added a new channel type. This channel type reads the XML header data out of input TIFF files and inserts them into the database. This channel is called "lookfor TIFF, get XML, insert document".
Version 2.1.1.8 - 2.1.1.7 - 2.1.1.6 - 2.1.1.5 - 2.1.1.3 - 2.1.1.2
Various changes to XML processor for IDOCs.
Thursday Feb 14, 2008
Version 2.1.1.1 - Update
Currently working through all of the different IDOC issues on the XML channel. At this point in time a version point release has been made since we have reached a stable point in testing. This version will remain in flux until a final specification has been reached with the customer. The IDOC channel setup has been added to the report server documentation page. Screen grabs and writing improvements are in progress.
Thursday Jan 31, 2008
Version 2.1.1.0 - Update
Reported an error in the case when multiple pages are embedded inside a single ICE document. All of the page data was being rendered onto a single page TIFF. Instead, it needed to be written out as a multipage TIFF with a repeating first page overlay, with subsequent data pages located and delimited using the COBOL 1 marker inside the ICE document. With a description and an sample file, we discovered why this error was happening, and coded the fix.
Wednesday Jan 30, 2008
Customer Issue
Client reports symptoms of a Report Server error, where the documents are being broken up across pages for the DMS and CSS feeds. These problems were previously addressed in version 2.1.0.9. The real problem is that the client is looking at documents that had been inserted with pre-2.1.0.9 ReportServer. The resolution is for the client to fix the old documents by reinserting them with the 2.1.0.9 version of the ReportServer. We conclude there was in interval of time when the DMS and CSS report inputs must have changed, but we did not notice.
Wednesday Jan 9, 2008
Version 2.1.0.9 - Update
Problem with DMS view skewing up the page.
When the COLD on ICE checkbox is on (this is a requirement for processing CSS and DMS) the ENDDOC field in the ICE document was getting replaced with an end of page character. This caused the routine that pulled the COLD file out of the ICE page to fail, because it was looking for the ENDDOC tag in order to do this. Now the page detection checkboxes do nothing when the COLD on ICE checkbox is on. We should simply use standard page detection for processing CSS and DMS. When COLD on ICE is turned on, a special page detection rule is automatically applied. This rule replaces the ENDDOC tag with the ENDDOC tag plus an end of page character.
Wed 1/2/2008 1:25 PM
Version 2.1.0.8 - Update
Problem with CSS and DMS overlays not lining up.
Compared CSS with DMS input feeds and noticed that "«« ACCOUNTING COPY »»" only appears in the CSS feed and not the DMS feed. This is used only at this client when the COLD on ICE checkbox is turned on in the parser panel to break multiple page letters into multiple page COLD documents. The DMS feed does not have this same anchor, so the COBOL marker "1/r/n" was used instead. According to the legend that Rafael sent to us, this sequence is suppose to delimit the beginning of a page.
Interpretation Of COBOL markers
// blank - space 1 line
// 0 - space 2 lines
// - - space 3 lines
// 1 - skip to the top of a new page.
