How to iterate over tables in an org buffer

I use org-mode for many things, especially tracking all the time I work on projects. To keep tracking as easy as possible, I use a table for each project. Each table consists of the fields: day, duration and comments. For me, there is no need to track certain tasks on a project, only the duration matters. Most of the projects have a lifetime of one year. I have to sync my tracking with our time-tracking-system each month, so I need an overview over all entries of a month ordered by the projects. At the moment I work on seven projects in parallel. You can imagine that the more all the tables grow, the more the org-file becomes unhandy. I don’t want to split up the file into several others to keep things simple. So I decided to write a lisp function that creates a new buffer and lists the entries of a given month of all tables. So let’s dive into some lisp code 😉

The tables in my org file almost look like this

There is a report-table in the same file. tot<IDOFTBALE> is the reference to the sum, and I use it to build the report-table. In further iterations, I will refactor the report so that this tag won’t be necessary for the future.

I want to exclude some tables from processing, so we store the names of those in a global variable.

Furthermore, we want to exclude some rows from processing, especially hlines, headers and summary lines. The following helper function tests if a given row should be excluded and returns true or false.

Next, we create our function body.

The function is declared as interactive so that we can set the month as a parameter in the minibuffer. The “M” in front of “Month” means, take an arbitrary text from the minibuffers input and return it as a string.

We can use the function org-table-map-tables to iterate over all tables in an org file. org-table-map-tables takes a function and invokes that on each table of a buffer.

org-table-map-tables operates on org-table structures. In my opinion, it is easier to work with Lisp lists. The function org-table-to-lisp converts the current org-table into a Lisp list structure. Keep in mind that it converts all rows including headers, hlines etc.

The function mapcar iterates over each line of a table and invokes a given function on each row. In Lisp it looks like that.

Let’s build the things together, that we have so far.

The structure above is all you need to iterate over all the tables and their rows. We will extend this structure to a full working example in the rest of this article, so you have something more realistic to play with.

First, we create our result buffer. Emacs Lisp provides the function get-buffer-create for this. get-buffer-create takes a string with the name of a buffer and returns that buffer, or it creates a new buffer with that name if it does not exist yet. We save the buffer handle a function local variable for future usage.

Each table in the result buffer should have the same name like in the origin org file. We can use org-element-property to get the name of the table.

org-element-at-point returns the org-element at cursors current position. We need not move the cursor ourselves, org-table-map-tables already does this job for us. In the next step we test if the table_name is in our ignore list, for that we have to convert table_name into a symbol first, the lisp function intern does the job.

To save the current tables name in our result buffer, we use the emacs lisp functions save-current-buffer and insert.

We use the same technique to copy relevant rows from our origin table into the destination buffer.

At the end, we open the new buffer in org-mode.

Here you find the complete code. Hopefully, I didn’t forget to copy something 😉

Let me know if this article helped you, or what things can be improved.

How to use cranyon backend plugin for org2blog

I use emacs, org-mode and org2blog for blogging in wordpress. For source code highlighting I want to use wordpress cranyon-plugin, but this plugin is not supported by org2blog. The following extension acts as a backend for org2blog and overrides the default behaviour.


For installation download the file and place it into your favourite lisp folder within .emacs.d directory. After that copy the following code into your init.el file.

Customize emacs leuven-theme – remove over- and underlines

Some people, including myself, don’t like the over- and underlining of headlines or sourceblocks defined by leuven-theme. Over- and underlining can be removed easyly. You only have to overwrite the matching attributes after loading the theme, this can be achieved with the set-face-attribute function.

I use use-package to load emacs packages, here is the complete code to load und customize leuven-theme