30th JANUARY 2019
Improved Visual Design: We refreshed the visual design of the Onedot application. It needed some color for the selected items, more space on wide screens and we put a shiny Onedot logo to the top-right corner so you do not get lost and always know where to return to.
Advanced Performance of Allowed Values: The performance of working with larger datasets involving lots of allowed values has been substantially improved. Moreover, when typing some text to search for a suggested value, only 100 values are now shown by default, hiding any less likely matches. Simply type again to find other suggested values if you don’t find what you are looking for.
Increased Data Blow Storage: New application stacks now make use of the new Block Public Access feature for S3 bucket, reliably preventing users or developers from ever allowing public access to our customer’s data.
23rd JANUARY 2019
Drag and Drop of Files and Folders: The existing Move action has been complemented with Drag and Drop! Files and folders can now be easily restructured by simply dragging them to the right place.
In a next step, the breadcrumbs of the folder view will also be converted into drop targets, allowing users to drop files to completely unrelated folders as well.
Fixed Dictionary Problem Rendering: The rendering of dictionary problems such as chains has been improved. When you change the offending rows to fix the problem, the error annotation now correctly disappears.
Sorting of Suggestions for Allowed Values: Sorting of suggestions has been fixed. The sort order now respects the sort order of the underlying dataset providing the suggestions, as expected.
10th JANUARY 2019
Hierarchical Allowed Values: This feature has been considerably extended! You can now also type a value in one of the columns further down the hierarchy, and all the upper hierarchies are automatically completed. Previously, completion only happened top-down, now it also works bottom up.
As an example, if you have a category hierarchy Electronics > Phones > Smartphone > Phone Cases, typing e.g. “Phone Cases” for a product “Superdry. iPhone X/XS Case” in would automatically complete the hierarchy up to the top level, as above. Expert users sometimes know their leaf-level categories by heart, so automatically completing the hierarchy bottom-up saves them a lot of work. On the other hand, typing e.g. “Electronics” on the top of the hierarchy would constrain lower levels of the hierarchy to be children of Electronics, such as Phones, Tables, Computers, etc.
Deleting Shared Files/Folders: The Onedot Software now politely asks the user when deleting a file or folder which is shared with a user from a same group.
This hopefully helps to avoid bad surprises if files or folders are not available for other users anymore, and the user deleting them is not aware of that fact.
20th DECEMBER 2018
Sorting of Folder Names: The new default sorting makes working with folders that use numbers to create a certain order easier. Folders frequently follow a structure and certain order, so we adapted that.
Allowed Values not Shown or Saves: In some cases, allowed values have not been properly shown, e.g. using latest Chrome
71.0.3578.98 browser. Everybody can happily update to the latest Chrome now.
XML Reading Improvements: A number of issues related to reading more complex XML files has been fixed:
- In case one XML element entry is missing, but the type of the XML element is inferred as a number, the Spark XML reader fell back to a String or undefined type. This is unexpected and has been fixed now.
- Previously, some nested XML elements containing attributes have been discarded by the Spark XML reader. We have fixed the problem and contributed the code back to the Spark XML Reader project of Databricks. Here is the corresponding commit.
- Other cases where attribute have been missing has been fixed.
13th DECEMBER 2018
Double-click to Resize Column: If you’d like to resize a column of the dataset view, just double-click on the resize icon (on the right side of each column) and the column width will change according to the counts in the columns. Works as expected in the dataset statistics view.
Color Your World: The colouring capabilities of the Onedot Software have been substantially improved. Now, you can go color entire columns and rows, not just cells.
We have also changed the general user experience related to selecting and manipulating rows and aligned it with the way you select and manipulate columns. Actions like Remove Column are now also available in the context menu when right-clicking on a row.
And on top, the visuals of the previous colouring user interface have been improved to match our visual design standards.
6th DECEMBER 2018
Improved Visuals on Safari and IE: A number of smaller issues has been fixed regarding the Onedot Software presentation on the Safari and Internet Explorer Browsers. Some background colours where missing, as well as separator lines which where rendered the wrong way. All good now, users still forced to use such browsers will rejoice.
Content Assist for Dictionaries: Editing and maintaining dictionaries has been taken to the next level: Whenever you add a new entry to a dictionary, you get a warning annotation on the domain if the value already exist in some mapping. Adding such a new entry would ultimately lead to an additional duplicate or fork. On top, you also get suggestions for range values when you add a new entry. The values suggested come from existing values of the dictionary range, assuming that they closely model the intended value distribution.
29th NOVEMBER 2018
Dataset Notifications: A new infrastructure to show multiple messages with different severities at once has been added to the user interface of a dataset. The message infrastructure is capable of showing a variety of warnings, errors or purely informational messages, at once, in a tool tip opening when clicking on a badge right next to the action buttons of the dataset.
If there is only one current message, the message is directly shown for easier access. Notifications have been implemented for various size limitations related to data set visualisation and editing. Also, messages are now shown as well if the Save button is disabled, explaining why it is disabled at the moment.
In the future, this functionality can be used to show arbitrary status information also related to the overall processing overview currently being drafted.
22nd NOVEMBER 2018
No Duplicate Allowed Values: The same allowed value shown multiple times in the drop-down was fixed, we did not like it.
Selecting all Columns: It is now possible to select all columns in a dataset, which comes in handy if you want to get feedback on a matrix-style dataset, e.g. for product family/attribute associations.
Automated Browser Testing: An automated browser test suite was provided with the overall goal to improve test coverage of the Onedot Software testing, but also avoid unwanted regressions in sales demos to our customers. The current test suite logs in with a user, loads the home page of the Onedot Software, sorts some folders and navigates to a dataset, sorts the dataset, clicks in the popup menu to choose some allowed values and navigates back via breadcrumbs.
“Onedot supports ZAGENO in our rapid market expansion by automating the vendor onboarding and making products more searchable through better product data quality. This ultimately leads to a faster growing global customer base and higher customer satisfaction.”
David Pumberger, CEO & Co-Founder at ZAGENO Inc.