Category Archives: SQL Blob Viewer

Why is SQL Image Viewer not displaying details of my blobs

If you are using a third-party application to upload images and/or files into your database, and SQL Image Viewer cannot identify the image or file type, then there’s a high probability that your application has modified the data.

Examples of such applications include the MAZE School Information System and the Financial Edge system by Blackbaud.  We had a user who had the following data stored in their Financial Edge database.  In SQL Image Viewer, the following is displayed:

SQL Image Viewer is unable to identify the data that’s stored in the fields (they’re actually PDF files).  If we look at the data using the SQL Image Viewer hex viewer:

we can see that the OLE wrapper (or the original source data) begins at offset 32.  This means that Financial Edge has added 32 bytes of data to the beginning of the original file, which is why SQL Image Viewer does not recognize the file format.

To identify and extract the data correctly, we need to skip the first 32 bytes, so that we only retrieve the original source file.  In SQL Server, we can use the following syntax:

Now, SQL Image Viewer is able to identify the file type correctly.

Ok, admittedly not everybody knows what an OLE wrapper looks like, or any of the other file headers, which is why if SQL Image Viewer cannot identify your blobs, please send us a couple of samples to analyze.  We need the data exactly as stored in your database, so to extract the data, please perform the following steps and send us the resulting files.

Select the column containing the unidentified blob data.

Right click the mouse button to bring up the context menu, and select the ‘Save item’ option.

Enter a file name, save the blob data, and send the file to us at support@yohz.com.

The same issue applies to Access OLE Export and SQL Blob Export too.  If these products cannot identify the file type because the original files have an additional header, they will be exported with a .bin extension.  Please send us a couple of those .bin files to analyze, or you can also use SQL Image Viewer to retrieve the data and follow the steps above to send us the samples.

SQL Image Viewer 9 and custom layouts

We just released SQL Image Viewer 9 with support for custom layouts.  Custom layouts allow you to customize how your data is displayed.  Take for example this default layout:

You see some textual values, and thumbnails of images from the result set.  Using custom layouts, the result could be displayed this way for production staff:

or in this layout for sales staff:

You can pretty much lay out every piece of data in any manner you want, depending on your needs.

Custom layouts can also display multiple frames from a single DICOM, TIFF, or GIF image.  Here is a custom layout showing frames from DICOM images:

Here is another image showing how custom layouts can be used for verification purposes.  In this example, the user wants to verify the scanned values against the actual values on the cheques, so a custom layout is used to place the scanned values right next to the cheque values.

Take a look at the videos describing custom layouts in detail:

  • how to create custom layouts
  • how to display multi-frame images
  • how to easily verify data from scanned images using custom layouts

Let us know how you’re using custom layouts, and how we can improve this further.  Do drop us a line at support@yohz.com with any suggestions or comments.  Thank you.

# Note that custom layouts is only available in the Professional Edition of SQL Image Viewer.

Exporting base64 encoded values in your databases

We recently had a prospect that had issues exporting her data using SQL Image Viewer.  She kindly sent a sample of the data, and it turned out that it was a base64 encoded string stored in a memo field.

So we got to work and now, SQL Image Viewer can decode base64 encoded data stored in memo fields.  It will recognize most image and binary file types.  For example, this is the result returned by Management Studio for a query that retrieves a base64 encoded PDF file:

 

This is the result returned by the same query in SQL Image Viewer:

 

You can immediately identify the type of data that’s stored in your database.  When you want to export the file, ensure that you select the memo field:

 

and also the Extract base64 encoded values in memo fields option:

 

For now, SQL Image Viewer will not inspect binary blob fields for base64 encoded data, as it wouldn’t make sense to store text values in a binary fields.  However, if you do find yourself in such a situation, do drop us a line at support@yohz.com.

Download a trial of SQL Image Viewer now, or buy a license from only USD 45.

Compressing PDFs in your database

→ This article refers to SQL Blob Viewer, which has now been renamed to SQL Image Viewer.  The techniques described in this blog is still applicable, as the functionality of the product remains the same.  Only the name has changed.

Recently, a user wanted to compress PDFs stored in his database, in order to reduce the overall size of the database.  He asked if we had any application that could do this.  Unfortunately, we don’t, but it got me to exploring the available options.

Turns out that PDF software development kits aren’t cheap at all.  Licensing can run into thousands of dollars, which isn’t feasible for us.  Open-source software is another option, which is what I finally went with.  In this case, I used Ghostscript, an all-purpose PDF toolkit, available at https://www.ghostscript.com/.

There are 3 steps to compressing PDFs in your database – extracting the PDFs, compressing or optimizing them, and finally uploading them back into the database.  We will use SQL Blob Viewer to first extract the PDFs, then Ghostscript to reduce the PDF size, and finally SQL File Import to upload the PDFs back into the database.  For reference purposes, these PDFs were created from document scans, so they have a 600 dpi resolution and are not optimized for PDF storage.  We’re running this example on Windows, but there is also a Linux version of Ghostscript, and both SQL Blob Viewer and SQL File Import will run on Linux using Wine.

 

Extracting the PDFs

Extracting PDFs from your database using SQL Blob Viewer is very simple – first write the SQL command to retrieve the PDFs.

We then export the PDF files to disk, using the primary key value in the ID field to name the exported files.  We do this so that when we upload the compressed files, we can use the ID value to update the correct rows.

If you have a lot of PDFs to export, you should choose to retrieve only the first few rows, to avoid loading the entire data set into memory.  After that, when you export the result set, the entire result set will be exported.  See this page for details on how to export large result sets with SQL Blob Viewer.

 

Compressing the PDFs

Now that we’ve exported the files, it’s time to use Ghostscript to compress the images found in those PDFs.

The easiest way to do this is to reduce the resolution of the images.  You can do this using the PDFSETTINGS option.  The possible values are:

  • /screen – converts to 72 dpi
  • /ebook – converts to 150 dpi
  • /printer – converts to 300 dpi
  • /prepress – converts to 300 dpi, color preserving

Depending on your requirements, you might want to test the various options to see which best suits your needs.  I took one of the exported PDFs, and converted them using each of the 4 options.  As you can see, the size of the PDF drops dramatically for all 4 options.

Here is the DOS batch script I used to convert the PDFs using the /prepress option (NOTE: Ghostscript options are case-sensitive, so you cannot for e.g write -PDFSETTINGS as -PDFSettings):

for %%x in (*.pdf)  do gswin64c.exe -sDEVICE=pdfwrite -dPDFSETTINGS=/prepress -dBATCH -dNOPAUSE -dQUIET -SOutputFile=”%%~nx_compressed.pdf” %%x

The options used are:

  • -sDEVICE=pdfwrite – this tells Ghostscript that we want to create a PDF file
  • dPDFSETTINGS=/prepress – this tells Ghostscript to convert all images found in the source PDF to 300 dpi resolution
  • -dBATCH -dNOPAUSE -dQUIET – these options indicate that the process should run non-interactively
  • -SOutputFile=’%%~nx_compressed.pdf’ – this tells Ghostscript how to name the output file.  Since we want to add a _compressed suffix, we first use the ~nx option to extract just the source file name without the extension, add the _compressed suffix, followed by the .pdf extension.
  • %%x – this is the source file name that matches the search pattern in the for %%x in (*.pdf) loop

Basically, this loops through all the PDFs in the current folder, converts all images in each PDF to 300 dpi, and saves the PDF with the _compressed suffix.

As you can see, the new PDFs are significantly smaller than the original PDFs.

 

Updating the database

Now, we need to update the existing record with the optimized PDF file.  We can do this using SQL File Import.  First, we enter the search pattern for the files we want to use i.e. those with the _compressed suffix.

Next, we need to map the columns.  Using the file name as the input value for the ID column, we need to:

  • extract the ID value from the file name
  • indicate that this value is a key field
  • indicate that this is an update process

We do this via the following script:

For the attachment column, we simply indicate to SQL File Import that we want to use the file contents.

Internally, SQL File Import will form the following UPDATE statement based on our script as follows:

UPDATE attachments SET attachment = :attachment WHERE ID = :ID

The test script shows that we have extracted the ID value correctly, and that the attachment column will use the contents of the files.

Now, we just need to run the script in SQL File Import, and our records are updated with compressed versions of the PDF files.

 

That is basically all you need to do if you want to reduce the size of the PDFs in your database.  The steps are similar if you want to process any of your blob data and update them in your database e.g.

  • resize images
  • compress files into archive (zip) files
  • process images e.g. add watermarks, convert to grayscale etc

SQL Blob Viewer and SQL File Import will handle the extraction and update process respectively.  You are free to use any external tools to process your images/files.

Exporting and extracting images and files from Microsoft Access databases

→ This article refers to SQL Blob Viewer, which has now been renamed to SQL Image Viewer.  The techniques described in this blog is still applicable, as the functionality of the product remains the same.  Only the name has changed.

Using OLE Object column types, there are 3 ways you can store images and files in a Microsoft Access database, or in a database (e.g. SQL Server) used by a Microsoft Access front-end.

  • embedding

Embedding involves creating the image/file directly using the associated OLE Server application.  For example, if I chose to embed a bitmap image, Access will open Paint for me to create my bitmap image, and save that image into that field.

  • insert from a file

If you already have the image/file you want to store in your database, you can insert the file into your Access database directly using the ‘Create from File’ option.  Access then copies the file into the field.

  • link to a file

Similar to the above, but this time, the image/file is not stored in the database.  Access just creates a link to the external file, much like a hyperlink in your browser.

If you store your images/files using the first 2 methods, then Microsoft Access adds additional data to your image/file, so that it knows which application to use to open that particular image/file.  This means that if you extracted the data from your database, it will appear different from your original file.

For example, let’s insert an image into an OLE Object field.  The image file is 18.7 Kb in size.

Once inserted into the Access database, the size has increased to 19.192 Kb.

The increase is due to the additional data added by Access.  If you now extract the data as is from the database, it cannot be opened by Paint because of the additional data.  The PNG data is preceded by OLE headers.

This is a common issue faced by Microsoft Access users – when they want to extract images and files from OLE Object fields to their original format, there isn’t a way they can do this easily.  Every image/file has been modified by Access.  You have to open each image/file in Access, then save the item to disk manually.

This is the reason we created Access OLE Export.  In just 4 simple steps, Access OLE Export will extract and export your embedded images and files to disk, stripping out the additional data added by Access.  This means that the exported images and files can be opened ‘normally’, using the appropriate application.

If you require more control over what is exported, SQL Blob Viewer is another option.  SQL Blob Viewer allows you to inspect your data before exporting them, but requires that you be able to write SQL commands.

There may be situations where both products are unable to identify embedded images correctly.  This may be because the registered OLE Server for that content type is not yet supported.  In these cases, send us (at support@yohz.com) a sample of the embedded data, and we will add support for that OLE server type.

SQL Blob Viewer 4 – email notifications added

→ This article refers to SQL Blob Viewer, which has now been renamed to SQL Image Viewer.  The techniques described in this blog is still applicable, as the functionality of the product remains the same.  Only the name has changed.

SQL Blob Viewer 4 was released last month, with a bunch of usability improvements.

One of it is the ability to send email notifications for the export process.  With the Professional Edition, you can schedule export jobs to run using the Windows Task Scheduler.  Previously, the job would run, and a log generated of the export process.  If there were any errors raised, you would only know about it if you inspected the log.

With email notifications, you can now receive a copy of the log via email.  You can also choose to have emails to be sent only when errors are raised.  The email options are configured in the Export Wizard.

logs_emails_01

Before email notifications can be sent, you would need to set up your SMTP mail settings first.

email_settings_01

 

Exporting images from OLE Object fields

→ This article refers to SQL Blob Viewer, which has now been renamed to SQL Image Viewer.  The techniques described in this blog is still applicable, as the functionality of the product remains the same.  Only the name has changed.

If you use OLE Object fields in your Access or SQL Server databases to store images, you know it’s convenient to just be able to click on the item and have Windows open the image using the registered viewer on your computer.  However, it becomes difficult to extract the images because Access adds additional OLE data to the stored images, thus changing their original form.

Take a Windows bitmap image,

and store it in an OLE Object field in a SQL Server linked-table.

The original file was 2,002,182 bytes in size, but has increased to 2,002,298 bytes when stored in the OLE Object field.  Access has added 116 bytes to the image.

and that the data stored in the field is different from the original bitmap file.

How can we then export this image?  One way is to open the table using Access, then double click on the data, and Access should then open the stored image in the registered OLE server for that image type, in our example Microsoft Paint.

However, this is a tedious process if we need to export a lot of images.  SQL Blob Viewer and Access OLE Export are two applications that we developed to easily export images and other data stored in OLE Object fields.  Exporting the items using SQL Blob Viewer or Access OLE Export is just a matter of writing the appropriate SQL query or selecting the right table.  SQL Blob Viewer is for users who are comfortable writing queries and want more control over how images are exported, while Access OLE Export is for users who just want to be able to select a table or write a simple query and export the images as is.

SQL Blob Viewer also displays a preview of the image when we query the table:

So if you have a ton of images that you need to export from OLE Object fields, give SQL Blob Viewer or Access OLE Export a try.

There may be situations where both products are unable to identify embedded images correctly.  This may be because the registered OLE Server for that content type is not yet supported.  In these cases, send us (at support@yohz.com) a sample of the embedded data, and we will add support for that OLE server type.

SQL Blob Viewer and SQL Image Viewer

So we have 2 very similar products, SQL Blob Viewer and SQL Image Viewer, and here’s why.

SQL Image Viewer was released 9 years ago, and over the years, it has accumulated a lot of code that is user-specific.  When we made the decision to create a 64-bit version of SQL Image Viewer, we discovered that it would be pointless to port those user-specific functions over too.  We did not want to leave our existing customers with a new version that did not have those functions, so we renamed the new product SQL Blob Viewer.

SQL Blob Viewer has almost the same feature set as SQL Image Viewer.  The most obvious difference is how it displays images – it does this by displaying the images together with the textual data –

while SQL Image Viewer displays images separate from the textual data (we’re still gathering feedback on whether users prefer the SQL Image Viewer way of displaying images).

SQL Blob Viewer can also embed images in exported Excel spreadsheets, has a user-friendlier interface to configure incremental exports, supports 64-bit versions of Access, and supports larger data sets with the 64-bit versions.  The plan is to improve on SQL Blob Viewer, while SQL Image Viewer will be updated only with bug fixes.  As both products were developed using different tools, it would be too much work to maintain 2 code bases.

In the near future, probably when we’ve decided if we should support the same views as SQL Image Viewer, we will release SQL Blob Viewer as SQL Image Viewer (new), as the name SQL Image Viewer seems to attract more traffic (I suppose SQL Blob Viewer is just a tad too technical).  When that happens, you will still be able to install SQL Image Viewer (new) alongside SQL Image Viewer (old), so you can still use both versions concurrently.  Existing users with valid licenses can request for a SQL Image Viewer (new) license.

I would encourage you to give SQL Blob Viewer a try if you have not already done so, and let us know what you think.  If you’re a SQL Image Viewer user with a valid maintenance license, contact us at sales@yohz.com for a SQL Blob Viewer license.