Captell - change log

Present in the 6.14.1.0 build this problem only occurs if a user tries to access a Captell report object directly using a URL without first having accessed the Web treeview or login form.

NOTE: This package contains only an updated Captell web installation, for a complete installation users must download the complete package 6.14.1.0 and this Captell web update. After installing the complete pckage, the Captell Web 6.14.1.1 installation should be used to upgrade the 6.14.1.0 installation.

Under rare circumstances the output from a pivot may not be correct when it is included in a document after tables are updated by the Captell service. This occurs when the pivot is using data from a query that includes a Captell parameter. When the service updates tables it automatically refreshes pivot meta data, however, when the query source of a pivot uses a parameter the query may not return any rows in which case the pivot will not be refreshed. This change corrects this issue by automatically refreshing pivot meta data before inclusion of refresh in a document.

A new facility under the Chart Edit menu "Add data table" enables the display of a table of the data values that are included in a chart.

Captell web now includes a clickable list of the 10 most recently accessed report objects for each database.

New WEB.CONFIG options have been provided that allows for specification of multiple Captell databases in the Captell web environment. The Captell web explorer will now show nodes for server, then database then the folders and objects.

More information is now made available in the Captell log when errors occur.

Report object exporting now produces XML files which allows bulk editing of report object definitions.

The "Order col" value controls the order of the columns in the output. The contents of the "Order col" must all be the same for each group of rows with the same "On columns" value. This change enhances the messages produced when this is not the case.

Several changes to underlying pivot processes have improved performance for pivots using a query as a data source by as much as a 50% reduction in elapsed time. Other minor changes have been implemented to improve reliability.

The table update processes have been updated to improve performance and reduce contention when multiple table updates are running consecutively. Source data is not loaded in an unconstrained manner to the SQL Server tempdb before validation, duplicate removal etc. and loading to the permanent database.

When a wildcard is used in the first position of an Excel or Text file table source type source name, the test connection and table update would fail to process the files correctly. This has been reslved in this release.

The task was correctly removed from the task list in the database however the task list display in the workbench was not automatically refreshed, this is now fixed.

Fixed. Changing and resaving a macro will now cause the metadata for any queries that include that macro to be refreshed automatically.

Fixed. This option is always available if the user is an Admin user or if the Allow Update users to release objects option is set.

After selecting a document from long list of document archives in the web the list would scroll back to the top forcing the user to scroll down again to select the next document, this has been corrected.

Publication parameters for Charts, Queries, Documents and Traffic lights are now correctly exported.

A new parameter used in a document, which is subsequently scheduled for document building using a driving query, where the driving query provides values for the parameter would not be updated until the Captell Service is cycled. This issue is now resolved by the Captell Service reloading it's copy of parameters before each document build.

Default text for an annotation point with a number or bullet list style was shown without the list attribute in the document preview. This has been resolved.

Fixed, Captell now tries to map to chart series title first, then if unsuccessful to chart series column name.

You now have the ability to include Captell documents in other Captell documents, and this inclusion ca be driven via a query. For example, you could establish a child document for a single server with a month and servername parameter, this document can include detailed information specifically for that server. You then establish a master document which includes the child, the inclusion is driven by a query that provides a list of the 10 busiest servers. The end result is a document showing details of the 10 busiest servers.

A new menu item, Toggle parameters hides or shows the parameter specifications for Charts, Traffic lights and Data displayed in the Captell web

Clicking a single column header in the web data display will sort the data by that column, note: this will involve re-executing the underlying query.

Changes to default web installation to install into the correct Application pool

The Captell web application must be installed to the Classic .NET AppPool application pool, using the Captell installer facility this is now performed by default.

Added information regarding setting the correct Application Pool for the Captell web application.

Test connection for Excel table source types failed when source file resided in disk root folder.

Under very rare circumstances a translation error occurs when a user connects to the Captell web

The error message "Some or all identity references could not be translated" is shown in the web browser when users who are members of non-current or invalid windows groups attempt to connect to the Captel web. This problem has been resolved by the system ignoring these groups.

Not all dynamic chart series attributes cloned when dynamically adding a new series

We have made some improvements to this facility, but further work is required to ensure all series attributes are copied to new dynamic series.

Chart dynamic series column names stored in the database CHAR and padded with spaces failed to be retained in chart definition

Dynamic series column names stored in the database as CHAR(n) failed to be retained in the chart definition when note all n characters were in use. For instance a column defined in the database as CHAR(8) containing a value of SYSA could not be used as a Dynamic Series Column name. This has been fixed in this release.

The By Column name field in a Dynamic Chart definition required [] when the column name included spaces or was a reserved SQL Server word. In this release the [] are automatically applied if required.

Setting chart web publication sizes in the chart definition would not affect the chart size when displayed in the web. Instead the chart was published to the database default chart size specifications. This has been corrected in this release.

Queries stored as Procedures become corrupt after database move or upgrade to SQL 2008 R2

Queries stored as procedures will become corrupt after a database is moved from one SQL Server instance to another or after upgrading SQL Server from 2008 to 2008 R2. This results because these queries rely on a linked server that is recreated during the move or upgrade. To correct this issue the TOOLS --> Refresh ALL Query metadata menu has been upgraded to recreate queries stored as procedures.

Previous release showed indented bullets in wrong order and the indentation was lost. This problem only occured in the preview and documents built from the web. Fixed.

Error when dragging a column name to the Ranges field in the query Mini bullet chart definition dialog

Fixed, changing the Axis Label Style property to PointValue in the chart properties Axis tab will ensure labels are correctly aligned.

Document build operations run from the web such as "Request a new document" or "Annotation preview" were quite slow when the IIS environment was installed on a 64 bit Windows server platform. This problem has been resolved.

Table updates on Disabled table run from Service attempt to show a dialog causing an error

Problems loading all data from an excel worksheet where columns contained mixed data types

The new OLEDB 12.0 32 and 64 bit drivers have different registry locations for controlling mixed data importing. A new Captell option has been provided to specify the correct location. The default for this options is determined based on SQL Server version and whether os is 32 or 64 bit, in most cases this should be sufficient however there registry keys can be in different locations.

CCL commands BUILD_DOC_NOW and UPDATE_TABLE_NOW returned wrong messages when supplied parameters are incorrect

Problems saving a query if it referenced another query that is saved as a procedure

Problems testing connection and saving wild carded Excel source type if no workbooks match wild card mask

Attempting to change a non nullable table column of type float, to nullable failed

Queries saved as Procedures with missing underlying tables showed a folder icon in the Captell Explorer

Browse button for the Service Working Directory Captell option doesn't work if the original directory doesn't exist

Workbench "flickers" when performing certain operations such as deleting tables, table data or queries or document builds.

Web annotation preview failed to show all annotations in a group if group was across more than 2 pages

Document build from the Build menu when building multiple documents failed to save the built document

New option in TEXT and EXCEL table source types to allow reprocessing already processed files

With this new option checked Captell will ignore the fact that the source file has already been processed.

Cutting and pasting Annotation points after previously deleting a point in the document higher than those cut caused the pasted AP's to loos any saved annotation text

Setting W (Wait) to NO causes the command to signal the caller immediately rather than waiting for the document build to complete. Setting the PA parameter to a ; delimited list of parameter names and values causes the build process to set those specified parameters before building the document.

Updated CCL commands UPDATE_TABLE_NOW to include W and to allow specification of multiple tables

Setting W (Wait) to NO causes the command to signal the caller immediately rather than waiting for the table update to complete. This release also allows specification of a ; delimited list of tables to be updated. The command starts a separate table update task for each table specified.

Changes to the Query based table source definition to allow specification of ISJOB

ISJOB changes the behaviour of the table update process so that queries stored as procedures that include calls to CCL UPDATE_TABLE_NOW and BUILD_DOC_NOW only run those processes once. With ISJOB specified all other normal table properties and operations are ignored.

The Captell Workbench menu VIEW --> WEB opens the default web browser pointing at the default Captell database (as specified in the Captell web configuration file), this release changes this behaviour to open the Captell database currently open in the workbench. This change also affects the Captell Web component in that the FRMCAPTELLLOGIN.ASPX page now accepts a Server and Database parameter, for example http://localhost/Captellweb/frmCaptellLogin.aspx?S=(local)\Captell&D=Captell_Samples. This allows a site setup a reporting home page with links to all the Captell databases.

Partitioning a table using a column that is the last column in a complex primary key would fail to partition the table. This problem has been resolved in this release. Also, changing the Table Requires Partitioning check box on the table partitioning dialog did not prompt to save the table definition on closure, this too has been fixed.

Including the control string #CAP_URL(text, url) in Captell query output which is subsequently used in a document will result in the text being rendered as a clickable hyper link either to an external web address or a document target or to another Captell report object.

All conditional formatting control strings including #CAP_CH (charts) and #CAP_TL (Traffic lights) can now be edited using simple dialogs. Using the Captell query definition dialog, right click the query text and either insert control strings or edit existing control strings. Conditional formatting control strings in Captell queries are now replaced with simple tags #CAP_CH, #CAP_MC_BAR, #CAP_FM etc which simplify the query, control strings can be edited by right clicking the tags.

Data (Tables, Queries etc.) included in documents are now formatted using default format options specified in the Captell Options Document\Tables dialog. The default formatting can be over ridden when inserting the data into the document or by accessing the inserted data properties; right click the table in the document..

UPDATE_TABLE_NOW and BUILD_DOC_NOW trigger table updates or document builds by talking directly to the Captell service rather than through the task list. These commands execute immediately rather than scheduling a task for execution. BUILD_DOC_NOW also allows specification of the built document name.

Changes to the method used to refresh document objects has resulted in a significant improvement in refresh speed.

A new popup menu appears when right clicking a table included in a Captell document, Change source data allows for the source data for the included table to be changed to a different query, table or subject.

This release includes both the 32 and 64 bit Microsoft Access Database Engine redistributables from MS Office 2010. These components are used by Captell to access text, Excel and Access data files. the New 64 Bit ACE redistributable allows Captell to load these data formats when run using SQL Server 64 bit instances.

The inclusion of multiple #CAP_xxx control strings in a single table cell is now supported

The #CAP_xxx control strings are used to embed charts, traffic lights, the new Captell mini charts and conditional formatting in Captell queries. These, when included in a Captell document or viewed through the Captell web show the included objects. This change allows for multiple control strings to be included in a single cell along with text and other document objects.

Ability to apply conditional formatting to table rows using #CAP_RF control string in first cell

The #CAP_RF control string can be used to set the background colour and font information of table rows generated from a Captell query. This control string is placed in the first cell of a row and can be conditionally constructed using a case statement in the SQL query.

Including a control string #CAP_BR in the first cell of a table row will cause Captell to remove all text and formatting from the row including the vertical borders (effectively creating a break in the table), the specified text will then be rendered in the first cell using specified formatting. This has the effect of creating breaks in the table and if the text is formatted with a heading style, headers in the table breaks which can be jumped to from the table of contents.

This facility inserts a field at the current location containing the number of pages in the document, as changes are made the field is kept up to date automatically.

Under some circumstances the table border widths of table/Queries etc. included in Captell documents were not remembered during table refresh operations. This has been corrected in this release.

The text box used to enter the Input SQL statements of table source definitions now accepts the TAB and enter keys. This allows for better formatting control when entering these SQL statements. This functionality has also been propagated throughout Captell where ever multi line textboxes are used.

Any Table, Query or Subject included in a document with a header row will have the header row automatically included whenever the table breaks across page boundaries. The page break header rows are created and maintained whenever the table is refreshed.

Custom colours added in any of the Captell colour dialogs will be saved and restored across Captell sessions.

The Insert Options dialog that pops when you drag a table, query or subject to a Captell document now has an option that allows for the selection of the columns that are to be included. This allows for a single query or table etc. to be included in different places in a document or documents with each inclusion showing different columns if required. It also allows for a filter to be specified limiting the rows included based on columns that are not necessarily shown in the document table.

An apparent problem in the MICROSOFT.ACE.OLEDB.12.0 and the MICROSOFT.JET.OLEDB.4.0 drivers prevent the above date time string from being read correctly, the drivers read it as 2010-05-04 05:59:59.997. The problem has been reported to Microsoft and we await a response, in the mean time, this change is to allow specification of the Microsoft ODBC driver for accessing desktop file formats such as EXCEL, Text or Access files. The ACE provider option on the General tab of the Captell Administrator now provides a "JET ODBC DRIVER" which when set will result in access to the above mentioned file formats being carried out using ODBC.

Update: Microsoft have stated this is by design (yes I know!!!) The issue arises because of rounding errors that occur when converting the higher precision values found in Excel, Text and Access to the less precise SQL date time values. Captell developments now recommends the use of ODBC drivers for accessing these source types.

On 6th May 2010 Microsoft released the new Access Data Engine for MS Office 2010, this includes a 64 bit OLEDB and ODBC drivers for Office file formats such as Text, Excel and MS Access. This change provides support for this new driver. The ACE provider option on the General tab of the Captell Administrator now provides a "MICROSOFT.ACE.OLEDB.14.0" options which is for the Office 2010 formats.

This change also introduces single threading during table updates that utilise OLEDB drivers, only the portion of the update that reads the external file is single streamed.

HOWEVER: a bug in the shipped driver prevents this driver name from working, until the bug is resolved by Microsoft you can must specify MICROSOFT.ACE.OLEDB.12.0 this provider will be the 64bit OLEDB driver if installed otherwise the 32 bit driver.

To utilise 64 bit ODBC drivers a new driver option has been included "JET ODBC DRIVER". This is the recomended driver option for both 32 and 64 bit SQL Server installations. See here for more details.

Captell mini charts are very small charts that are rendered directly in query output either to a Captell document or to the Web, that is they don't have

any underlying Captell chart object. Click the thumbnail for a small example. The Captell mini chart facility allows control strings to be generated by a query, these instruct the Captell document and Captell web to render the charts. Currently supported are mini bars, lines, pies and bullets.

Mini charts do not have any labels or headings etc, they provide limited formatting capabilities. They are ideally suited for use in a dashboard style report to present a large amount of information in a small amount of space. Click here for a preliminary sample document.

This problem occurs when the currently defined list of users is not large enough to require scroll bars, then a new user is added which makes the list too large to be displayed completely, the scroll bars would become visible but not active. This problem is now fixed.

A new table source type "Existing table in database" has been added that allows for a table to be created, loaded and maintained from outside the Captell system, this table type is effectively a place holder to show the table in the Captell explorer and therefore allows use of the table in Captell queries and other Captell report objects. The underlying table must adhere to Captell table naming conventions i.e. it must contain at least one folder component [foldername\tablename].

Minor improvements to elapsed times when updating tables with source types of Excel, Query based, Jet or ODBC. Under some circumstances the time taken to determine source data schema has been reduced.

If the SQL text is shown in the Query definition window, pressing the F5 key will (in a similar fashion to SQL Server Management Studio) executes the query and show the results

This enhancement allows Captell queries to be saved as procedures, which allows queries to be created that make use of temporary tables and that can define and access SQL Server cursors. Essentially a Captell query saved as a procedure can do anything that a SQL Server Stored procedure can do with the additional benefit of being accessible via a FROM clause of a SELECT statement. e.g. You can write a query that utilises temporary tables, cursors and calls other procedures to manipulate data, the final step should be select * from a temporary table. The data provided by this query saved a procedure can be consumed in other Captell objects such as Charts, Documents or can be utilised in other Captell queries by simply selecting from the query name. SELECT * from querysavedasprocedure.

This change also gives access to other SQL Server stored procedures by simply including EXEC storedProcedureName as the query text.

Dynamic chart groups can be dragged to a document at which time all charts in the group are rendered to the document. Prior releases had a bug that prevented single charts from the group being used in a document.

Release 6.12 introduced a bug that prevented the Manage Access explorer menu item functioning properly.

Captell 6.12.0.1 includes no new functionality only performance improvements, and one minor bug fix specifically:

Improved performance and accuracy of query meta data refresh when editing queries. Any queries that have a dependency on the query being edited will have their meta data refreshed when the query is saved, this process now occurs in a stored procedure on the database and thus eliminates round trips between the Captell workbench and the SQL Server.

Improved performance of Captell explorer refresh. Only one trip to the database is made to refresh the Captell explorer. Improvements in access control have also significantly reduced the time taken to refresh the explorer.

Improved performance and accuracy of pivot meta data refresh when editing underlying queries or changing parameter values that are used in underlying queries. This functionality is now carried out in a Stored procedure on the database thus eliminating database round trips. Previous versions of Captell did not refresh the pivot meta data when a parameter value (used by an underlying query) was changed. This resulted in pivots not showing the desired results when a parameter value change reduced or increased the number of data values available for use as columns in the pivot.

Improved ability of Captell web to display output from #CAP_SET_STATUS and #CAP_FM functions.

For small sets of data usually provided by a user in your network, the Captell Subject Data Entry facility provides an easily established dynamically built data entry screen delivered to the user via the Captell web. Data entered via Subject Data Entry is stored in a common table, each subject area defined has a separate view automatically generated that allows the subject data to be consumed in other Captell report objects.

Captell documents can now be published as dashboards, which results in the document being rendered on the fly into PDF for viewing in the Captell Web facility. The dashboard can be simply laid out by dropping Captell report objects into tables. Any element allowed in a "normal" Captell document can be used in a dashboard, including Charts, traffic lights, plain text, text provided from a query and table and query output.

Parameters used to drive the content of report objects can be published alongside the dashboard allowing users to refresh the dashboard simply by changing the parameter values.

The DBO.CAP_SET_STATUS function accepts a value column and a threshold, it compares the value to the threshold and returns a string made up of the value along with a character which is set (including its colour) depending upon whether the value is greater than, equal to or less than the threshold. The output from this function makes an excellent summary table for documents. The character can be any character for any available font including Wingdings.

Previous releases were very restrictive when editing manually created table content in a Captell document. Most of these restrictions have been removed.

Report object in-use locks will be ignored if the current user is the user who holds the lock. The in-use lock can be ignored from the message indicating the object is in use; you no longer need to go back to the Explorer to release the lock.

Table and query output can now be inserted into manually created document tables

Output from queries and tables can now be dropped into manually created tables in the Captell document. This allows for much finer control over table/query output in a document. Previous releases have allowed charts and traffic lights to be included in manually created tables; with this inclusion charts can be shown with their respective data side-by-side.

This license type is no longer supported, Desktop and Enterprise (perpetual and annual) are supported.

The recent release by Microsoft of their BETA 64 Bit OLEDB ACE drivers (see download here) has prompted this change to allow for specification of different OLEDB drivers either MICROSOFT.ACE.OLEDB.12.0 or MICROSOFT.ACE.OLEDB.14.0. This specification is made by the Captell Administrator using the Administrator application. NOTE: at time of writing as these drivers are BETA we urge caution if considering their use in a production environment.

Under some circumstances editing a document caused an out of memory condition. This was traced to large image sizes in documents built on older versions of Captell. This has been rectified in this release by automatically changing the image resolution when opening a document for editing.

Table, Pivot and Query viewers and editors now has a maximum row output count, set to a default of 20000. Once this limit is reached the transaction will be stoped. This is to prevent memory conditions when "accidentally" trying to view several million rows of table output. Changing the Maximum rows value to 0 will display all possible rows.

Attempt to release MUTEX not owned by caller error when updating an Excel sourced table

This error was occasionally thrown when updating multiple excel workbooks in parallel. It has been resolved.

The new table disposition Truncate option failed if the table didn't already exist. This has been resolved.

This removes all previously discovered SQL Server instances from the list. <Browse for more> discovers new instances.

Long parameter values are now shown as tooltips when hovering over them in the parameter drop down list.

In previous versions parameters displayed in the Captell web in alphabetical order. In release 6.12.0 the order can be manually set when adding parameters to the Publication dialog of the various report objects.

Subordinate parameters are those that are dependent upon the one being defined. For instance DIVISION, BRANCH and SECTION parameters, BRANCH and SECTION are subordinate to DIVISION. Specifying subordinate parameters causes their selection lists to be correctly redrawn when a higher parameter value has changed.

A query that only included --# variables if accidentally saved as a view the query editor would hang. This has been resolved.

If a section of query text from a query containing a #INCLUDE statement was commented out using the Comment out option and the #INCLUDE contained multiple lines of query text, the second and subsequent lines of the included text were not commented out in the resolved query.

Many changes made to improve overall performance by reducing the number of round trips to the database server.

Drag an object from the Captell explorer to the flow diagram and refresh the page and all is well. Now delete the object from explorer and refresh the flow diagram and an error occurs.

Table updates from a common ODBC DNS will now single thread. In previous releases Captell could start multiple table updates from a common DNS which could result in failures.

New wizards have been added to provide summary queries and tables containing basic statistics for a source table.

The Excel source type now warns when the source workbook is not present during definition.

Document build doesn't produce disk output if the document name contains a : character

: characters found in the document name (Usually included from a driving query) are now replaced with spaces

Options Working directory browse button failed if previous directory didn't exist.

Default annotation text from an annotation point with a number of bullet style didn't include the bullet or number.

Better messages when a user connecting to Administrator specifies password with Windows account

Previous releases would not accept time value from the monthly definition, instead it used the time value specified as part of the daily schedule. The monthly schedule now correctly uses the correct time value.

Table of contents now searches for any paragraphs formatted with heading 1, heading 2, heading 3 or Heading 1, Heading 2 or Heading 3 styles.

Up to 5 desktops each with a single SQL Server Express instance each supporting a single database (4gb in size) are allowed with a desktop license. Each desktop in the license can access the databases on any of the other desktops.

The desktop license does not include access to the scheduler side of the Captell service; however, Run at Server can be manually performed.

An error introduced in Captell 6.11.0 prevented the correct display of area lines on area charts. This has been corrected in this release.

If a query produces output that includes the control string #CAP_CH(chartname,width,height) Captell will replace the control string with the chart when the query is used in a document or viewed on the web.

The Edit --> Targets menu of the Captell document definition window now allows easy removal of TOC document targets.

Dragging a traffic light or a chart to a query definition window will drop the correct #CAP_CH or #CAP_TL control string.

A new Truncate menu options appears when right clicking a table in the Captell explorer. Truncate removes all rows from the table leaving the table structure in place. Delete table data deletes the underlying table without deleting the table definition and Delete Definition removes all the data, deleted the table and deletes the table defintion.

The web based annotation editor pops a spell checker when closing, the page will not completely close now until the spell checker window is closed first. The spell checker window now stays on top of the annotation window.

Annotation word count incorrectly accumulates every time the annotation is saved

The Delete column drops the column from the table unless the column is used in the Primary Key or for data culling or partitioning.

%CURRENTSERVER% and %CURRENTDATABASE% accepted in the SQL pass through defintion

These are replaced with the current server and database when the table is updated.

Previous releases would remove document definition node from the Captell explorer when the last document archive was removed. Note: The document wasn't deleted from the database only removed from the explorer. Clicking Refresh correctly redrew the explorer showing the document. This has been corrected in this release.

Under some circumstances when opening a document an out of memory error was thrown. This has been traced to large image sizes when the document internal text was expanded for editing. this has been corrected in this release.

This has been necessary for environments where the Captell service runs in a different time zone to the Captell database.

Under some circumstances (usually when closing a number of edited queries simultaneously) the Captell explorer could erroneously show the same object several times. This was a result of the explorer being refreshed by each closure. Captell now single threads explorer refresh operations. And trigger to refresh the explorer while a previous refresh is underway will be ignored.

Under some circumstances an overflow error resulted from the CAP_CORCOE function. This has been resolved in this release.

If a URL referring to a Captell report object specified the S and D options Captell will automatically attempt to logon to the specified database and display the object. If logon fails the Captell web logon window will be displayed.

An error was introduced in the final production build of 6.11.1 that prevented hyperlinks from being defined, this has been corrected.

Under certain circumstances table update summary table processing would trigger a table update on the base table (the table initiating the summary table updates). This resulted in two tasks in the task list for the same table, the initial task waiting for summary table updates to complete and a new task waiting for the initial task to complete. This has been rectified, summary table updating will no longer trigger a table update task on the base table.

An error was introduced in the final production release of 6.11.1 which prevented annotation points from having a hyphen in their name, this has been corrected.

Issues resulting when different versions of the Captell workbench are used to access a Captell database have prompted this change. The Captell Administration application can now be used to set application version information into a Captell database. Any workbench client of a different version attempting to access the database will be rejected.

To prevent display of this message when users of databases prior to 6.11.1 open the database in the Captell Administrator application the Security tab has been disabled until the database is upgraded.

Under certain circumstances an error was thrown when attempting to create new databases using the Captell Administrator application, this has now been rectified.