A Plone page with an editable table as main content
A new Plone content type similar to the standard Page but with a table as main content.
Table of contents
- How to use
- DataTables integration
- Searching the table
- Other products
This product want to give to site members a simple way to manage a page with a table inside in a collaborative way. To be more precise, it’s focused on contents where the table is the main scope of the page.
The only other option is to create the page using the WYSIWYG editor (like TinyMCE), then leave to editors power to modify it, but:
- using TinyMCE commands for table is not so easy (users sometimes mess up your pre-defined layout)
- you can’t prohibit users to delete or change rows added from other users, or adding new column you don’t want.
If you need to beat those limits but you still simply need a Plone page, this product is probably what you need.
If you need to store a huge amount of data, you should probably look for other solutions.
After installation you will see a new addable content type: the Page with Table.
Some fields of this new content types are very similar to Page ones, although the “Body Text” field is splitted in two separated sections (text before and after the table).
The most important field is “Columns”, where you can define the column structure of you table.
For every column you can define some information like header’s content and other description, but you must also define the type of data in the column.
You can add as many columns as you want; users that will fill your table won’t be able to change what you have defined.
Configuration is not changing anything in your layout, but users with Contributor role on this document will see a new tab: “Edit table”.
When accessing the “Edit table data” view, users will be able to add new rows to the table and edit their own rows. The form given to the user is generated using the configuration options that the document creator defined before.
- Every added row is put at the end of the table or at the end of section (see below)
- Every Contributor is able to edit or delete his own rows
- Users with “Editor” roles are able to edit or delete all rows
- Users with “Editor” roles are able change row order
When switching back to main document view the generated table is part of the document body text.
Users with power of configuring the table can also add a special type or row: Label. Apart the UI changes, labels break the table in groups of logical rows: every group start at the position of the label at end at the next label (or at the end of the table).
If one or more labels are used, contributors will be able to add new rows at the end of the section instead of adding only at the end of the table.
In the table configuration there’s a field named “Criteria for adding new rows”. Changing this value from “At the end” (default) to “At the beginning” will change the adding method: new rows will be added at the beginning of every groups.
Data stored in the table can be downloaded, and optionally you can display a download link also to page visitors (activate the “Show download link for data” inside “Settings”). When the download icon is used in the “Edit table” view, downloaded data is compatible to the upload CSV feature described above (columns ids are used instead of titles, contents uids instead of URL to referenced contents, …)
Contributors can also upload data using a CSV file. The file must provide a row with column ids defined in the configuration. Columns with an unknow id will be ignored.
If the configuration has not already been defined, all CSV headers will be used to quickly init the configuration (but columns types will always be “String”).
Table Page is distributed with a know set of columns. Right now you can choose from those types:
- A simple line of text, the most common (and default) type.
- A textarea, for saving more text and take care of carriage returns. Cached for 12 hours.
- Still a simple line of text, but user must choose it from a vocabulary you will define in the “Column configuration”.
- A link to a file in the site. Cached for 1 hour. See below.
- Same as File above, but for multiple files.
- A link to an URL, or an internal site document. It use Plone reference browser native widget. Cached for 1 hour. See below.
- An e-mail address.
- A string in numeric format.
- A string in numeric format, but will be rendered as a monetary value, with locales settings.
- A column that will display a value based on a computed TAL espression you must put in the “Column configuration”. For this reason it will not be putted in the edit row form. Not cached by default but can be configured. See below.
Adding new type of column is not hard (for a Plone developer), but remember to stay simple: we don’t want to rewrite PloneFormGen from scratch!
Some of the columns above enable a persistent cache. This can be needed for large tables, where a lot of those columns can slow down the page.
Cache is automatically invalidated when the column is modified, however is possible that invalid data is still shown in the table. For example: you create a link to a file, so displaying it’s title, but meanwhile an editor changed the title of the file.
Columns of type file(s) are the most complex.
When adding or editing a row the user is able to upload new files, creating a new Plone File content, or selecting existing files from the site. In both cases permissions matters: the user must have permisson of adding new file in the storage folder or see it. The storage folder is configured by the document creator.
When rendering the table, a link to download the file is displayed.
When writing the TALES expression to be used in computed fields you can access general vars like:
- The current page with table
- The Plone site root
- The index of the current row
- The current row. using this you can access data taken from other columns in the same row.
The row var in the most powerful: based on the type of column you are referencing, you can read different data. For example: accessing a File, Files and Link column, you can read information of the referenced object.
(show the title of the file_column column in the same row)
(show the title of the first file in the files_column column in the same row)
(show the link of the link_column even it’s an internal link or an absolute ones)
Even if this column normally don’t implements any cache, you can specify a custom cache by defining an additional configuration line in the “Column configuration” field.
Just write something like…
…to cache column’s result for an hour.
TablePage has a soft-dependency on DataTables; if the jQuery plugin is installed, the table view try to use it for getting some new features like:
- filtering/searching data
- sorting by columns
Apart the live search filter that came from DataTables integration (see above) you can rely also on advanced search features. This will give to your users a search form automatically generated looking at search configurations.
The search feature is based on a ZMI tool: tablepage_catalog really similar to the same catalog used by Plone for it’s search engine. While some UI configuration are possibile through Plone, a ZMI access to that tool is required.
For every “searchable” column you have defined, you can create a field in the search form, customizing the label and the helper text. Plus, you can define one or more columns as searchable in full text search.
When you users will perform searches from the table view, only rows that match the search will be displayed.
The widget displayed in the form depends on the catalog index you user:
- for a ZCTextIndex you will get a text input
- for a FieldIndex you will get a selection on all possible values
Names of the indexes must be equals to columns ids.
You have a lot of limitation:
- no label support right now: do not use search feature if you have label (this will be fixed)
- no other kind of indexes are supported right now
- you have one catalog, so you must handle (or avoid) columns ids used in more that one page with table
There are at least two other products for Plone that are focused on table generation:
- This product is focused on the editing part (and the use of DataTables jQuery plugin is nice), but it dowsn’t work on Plone 3 and you have no way of limit the power of users on the table.
- Very powerful, modular and extensible. It’s using PloneFormGen as table configuration and can store a lot of data. Unluckily it has a lot of dependencies and it won’t run on Plone 3.
This product can be used with al version of Plone from 3.3 to 4.3.
For Plone 3.3 you need some special configuration like:
- A custom branch of DataGridField where we backported some new features from 1.8 branch
- Available table styles are taken from TinyMCE configuration, so you must use it instead of Kupu
- No versioning support is available
- No friendly installable DataTables product is available for Plone 3, so you probably can’t use it
Developed with the support of:
All of them supports the PloneGov initiative.
- Fixed a bug that break CSV export when computed columns are used [keul]
- Added minimal Link colums diplay prefs (a fixed link text or icon) [keul]
- Fixed wrong column configuration cache [keul]
- Fixed appearence of “No rows” section on Plone that are not using plone.batching [keul]
- Enable DataTables only if we have some rows to show. This fix viasual issues with some layotus [keul]
- Link column: put the external value for rel for external links and not for internal ones [keul]
- Added search features [keul]
- Multiple tables view was unreachable on emtpy tables [keul]
- The jquery.dataTables.rowGrouping.js plugin is disabled by default [keul]
- Multiple multi-files columns in the same table was not working [keul]
- Styles fixes: main column (HTML) label is a little bigger that default Plone form labels [keul]
- Prevent new label from load a wrong default text [keul]
- Do not display empty icon in link column [keul]
- New “insertType” configuration (new row at the end or beginning of groups) [keul]
- New column type: “Computed” [keul]
- Fixed a problem with link-like columns and cache. Do not return object absolute_url because a backend URL could be cached. Instead use the resolveuid URL and run table through portal_trasform when in view. Drawback of the approach: when editing the table’s URLs still use resolveuid [keul]
- Added batching/pagination [keul]
- Multiple tables view was not properly display HTML [keul]
- Added caching for rendered columns. This will speed up a little/lot table rendering [keul]
- Show/Hide command now act also on page header and footer (Zen Mode!) [keul]
- The unique validator was preventing record update [keul]
- Monetary column will pad the final zero in less that 2 decimal are supplied (123.5 will be 123.50) [keul]
- Fixed a bug that break link columns when the linked content is no more [keul]
- Fixed error when validating old rows, created before version 0.5 [keul]
- Do not use the HTML 5 number type anymore because of Google Chrome stupidity [keul]
- Fixed a Python 2.4 bug in interpreting CSV format [keul]
- Do not fail the whole import procedure if a CSV row is missing some columns [keul]
- Fixed error when editing old rows, created before version 0.5 [keul]
- The import from CSV form can be used when no configuration has been given. A basical configuration will be guessed by columns headers [keul]
- Select colum now enforce vocabulary values [keul]
- New column type: “Monetary” [keul]
- When exporting in CSV, always quote data. This prevent some fancy Excel/OpenOffice interpretation [keul]
- Column validator can be executed also when importing from CSV [keul]
- Soft dependency on jQuery DataTables plus “Row Grouping Add-on”. This add new features like live-search in table, batching and colum sorting. [keul]
- Fixed critical error in the “Files” column; when selecting existing file the column id was ignored [keul]
- Added new feature: registering validators [keul]
- Added validator for required field [keul]
- Added validator for unique field [keul]
- New field type: “Email”, for inserting an text in e-mail format [keul]
- New field type: “Numeric”, for inserting an text in numerical format [keul]
- Added uninstall profile [keul]
- Fixed bug in finding duplicate rows when importing from CSV (close #1) [keul]
- Do not display selection checkbox if I can’t delete a row [keul]
- Raise lifecycle events properly when creating files [keul]
- New field type: “Files”, for uploading a set of files to be rendered in the same cell [keul]
- Labels inside the table are now supported [keul]
- New view for displaying data on multiple tables [keul]
- New field type: “Link”, for inserting an URL or an internal reference [keul]
- CSV export done by backend get UUIDs when applicable [keul]
- CSV import now validate data: do not import every text you read from the file [keul]
- CSV import now transform URL/path to valid content uuids [keul]
- Different versioning message when a row is changed or modified [keul]
- Added missing versioning attempt when using CSV upload [keul]
- Fixed a performance/security problem: data inside text cells were transformed to HTML without any check (and this was also really slow) [keul]
- Can now delete multiple (or all) rows [keul]
- CSV import is not importing anymore inside wrong colum when an unknow header is found [keul]
- Fixed missing translations [keul]
- Do not display “download as CSV” for empty tables [keul]
- Added an option for choosing when display headers [keul]
- Handle loading of duplicate file id: file is not loaded twice but same reference is kept [keul]
- Do not display “Edit table” or row’s commands if no configuration has been set [keul]
- fixed encoding error on columns headers [keul]
- fixed encoding error on editing rows [keul]
- Fixed UnicodeDecodeError problem with non-ASCII chars [keul]
- Initial release