guypursey / ahem

An idea for a blogging workflow.

ahem

v0.1.1

This repository represents elements of a workflow I use for blogging.

Currently, my blog appears on Scriptogram at http://scriptogr.am/guypursey

Dependencies

This workflow relies on Git for versioning.

Set-up

Currently this workflow depends on posting to Scriptogram. You can either:

• Comment out the relevant line in /code/publish.sh, or:
• Create three text files called scrpakey.txt, scrpuser.txt and scrpname.txt in the folder /userdata/scriptogram containing the application key, user ID and username of your account respectively. Further details are given in the section on "Absent Files" below.

Usage

• Create a new draft by using /code/newdraft.sh followed by name of folder to appear in content/drafts/.
• Edit the main .md file in the newly created folder to start writing your post in Markdown.
• Precede relative links with /$/.
• Precede local image references (for images in the /content/images folder) with /!/.
• Add a keyword to each line of tags.txt in your newly created draft folder.
• Commit amendments to your post and track your progress. Each draft folder is its own Git repository.
• Publish the post using /code/publish.sh following by the name of the relevant folder, which will move your folder from ./drafts to ./archives and create a copy of the post in ./posted.

You can optionally set the date of publishing by putting an ISO format date into a file called date.txt in your draft folder. If you don't do this, one will be creating automatically when you publish.

Example

While in the root of the repository:

/code/newdraft.sh new-blog-post
echo "# My new blog post" >> /content/drafts/new-blog-post/new-blog-post.md
echo "Testing this out." >> /content/drafts/new-blog-post/new-blog-post.md
echo "testing" >> /content/drafts/new-blog-post/tags.txt
echo "2015-03-08 12:04" > /content/drafts/new-blog-post/date.txt
/code/publish.sh new-blog-post

Folders & files

Here is the directory structure:

/.
|__ /code
	|__ /picasa
		|__ authenticate-as-device.sh
		|__	poll-for-access-token.sh
		|__	post-image.sh
		|__	refresh-access-token.sh
	|__ /scriptogram
		|__ autoscrp.sh
	|__ archive.sh
	|__ newdraft.sh
	|__ publish.sh
|__ /content
	|__ /archives
	|__ /drafts
	|__ /images
	|__ /posted
|__ /userdata
	|__ /picasa
	|__ /scriptogram
|__ .gitignore
|__ README.md

There are also README files besides the root-level one in some of the folders which aren't listed above, so as to make the directory structure clearer.

/code

This folder contains all the scripts that take care of this model of writing, versioning and publishing content.

/code/picasa

This folder contains any code dealing with publishing images to a specified Picasa account.

/code/picasa/authenticate-as-device.sh

This script deals with authenticating this code with your Picasa account.

/code/picasa/poll-for-access-token.sh

This script deals with obtaining an access token for publishing to a Picasa account.

/code/picasa/post-image.sh

This script deals with posting an image to a Picasa account and storing the URL returned in an image map (csv file) in the userdata/picasa folder.

/code/picasa/refresh-access-token.sh

This script uses the refresh token stored in userdata/picasa to get a new access token. Essential for one an access token expires.

/code/scriptogram

This folder contains any code dealing with posting to Scriptogram via its API.

/code/scriptogram/autoscrp.sh

This is the script which pushes a post to Scriptogram, creating or pulling in the metadata as required.

/code/archive.sh

This is the script which moves a folder from drafts to archives and renames it, as well as the file within, for when a post has been published. A copy of the file by itself is also made in posted.

/code/newdraft.sh

This is the script which creates a new draft folder with an empty .md file in it ready for writing. The draft folder is initiated as a Git Repository.

/code/publish.sh

This single file can be used to archive a piece and push it out to Scriptogram in one command.

/content

This folder shoul contain all the content for your project/blog.

/content/archives

This folder will likely remain empty, unless I decide to share my writing process more generally. Locally, it contains Git repositories for all the posts I have published.

/content/drafts

This folder will also likely remain empty for the same reason. Part of my publishing process is the transfer of a folder from /drafts to /archives so that I know what's out there and what still requires work. I can still make changes, corrections, etc. But it keeps files that I know are work in progress separate from files I know I've put out there.

/content/images

This folder is a placeholder for images. It contains a README and can be used to house images to be used in posts. It should contain a file called imagemap.csv containing information about where images are hosted online. Further information below.

/content/posted

This folder will contain all the Markdown files that make up my blog (as soon as I've worked out an appropriate license). It's a duplicate of the content on Scriptogram without any of the bumpf.

/userdata

This folder is for storing account details which should not be shared more widely (access tokens, user and client IDs, client secrets, and so on).

/userdata/picasa

Account data for Picasa. Should not be versioned or stored remotely; keep local.

/userdata/scriptogram

Account data for Scriptogram. Should not be versioned or stored remotely; keep local.

.gitignore

Ensuring local files with all the drafting and editing that went into them stay local and hopefully helping to ensure that account details stay local and private.

README.md

You are reading this file now.

Essential but absent files

If you want to use the same approach to your blog, with Scriptogram, you will need:

• A Scriptogram account.
• An "application" added to the Scriptogram account that allows you to post.
• Two files described below.

If you want to use images in your posts, you can store them in the /images folder but you will need to map them with a file called imagemap.csv which is described below.

You can use a Picasa account to store your images and have the image publishing and mapping taken care of automatically.

/userdata/scriptogram/scrpakey.txt

This should contain the application key for Scriptogram which can be got by adding the "application" mentioned above. It should contain nothing else, no spaces, no line breaks, nothing.

/userdata/scriptogram/scrpuser.txt

This should contain the Scriptogram user ID, which you should find under "Settings" when you login to Scriptogram. It should contain nothing else, no spaces, no line breaks, nothing.

/userdata/scriptogram/scrpname.txt

This contains the actual username which appears in the Scriptogram address, used for the purposes of crosslinking.

Additional but absent files

You can use a Picasa account to store your images and have the image publishing and mapping taken care of automatically.

/userdata/picasa/_client_id.txt

Obtained by setting up a project in the Google API console.

/userdata/picasa/_client_secret.txt

Obtained by setting up a project in the Google API console.

/userdata/picasa/_device_code.txt

Obtained by setting up a project in the Google API console.

/userdata/picasa/imagemap.csv

This file should contain the filenames of any images in the /images folder where it sits, followed by a comma and then the address of where that image is hosted online. Each file and address pair should sit on a separate line. If using Picasa (as the scripts are currently set up to do), this file should be handled automatically.