booherbg

How to build Golang windows/arm static binaries from linux

Alternate title: Cross compiling Windows/Darwin/Linux amd64/386/arm all from linux

After fumbling around trying to figure out the go toolchain and cross compilation configuration, I ran across the wiki page on Go's homepage. It's super helpful, and worked out of the box. I'm including the necessary scripts here in case they get changed or lost, and we can help Google find it (since it's the first real source I've found that "Just Worked"). http://code.google.com/p/go-wiki/wiki/WindowsCrossCompiling

booherbg

For my own reminders...

General

• Go is fun, like python fun.
• Structs are fun, easy to use
• maps are fun, easy to use
• Pointers, references act as you'd expect them to
• Duck typing, static typing mix
• Feels expressive, like python. Compiled and fast, like C++

booherbg

The documentation online can be confusing, especially with config paths from old files floating around.

1. Download an image file for RetroPi (Ideally 2.5+)
2. Unzip the image file, plug in your SD card (likely /dev/sdb)
3. dd if=retropi.img of=/dev/sdb (this will erase your entire SD card)
4. Optional: increase the size of the new partition with gparted

System Setup
1. Add your ROMs to ~/RetroPi/roms/snes (which will cause your EmulationStation to now show the system in the menu). Be sure to unzip them.
2. (Not needed in RetroPie 2.5) Update your controller config, either by running ~/retropi_setup/retropi_setup.sh, or copying a known config to /opt/retropie/emulators/RetroArch/USB,2-axis8-buttongamepad.cfg

booherbg

During install, my /dev/sda was not recognized (even though fdisk -l shows it). I did, however, get prompted about enabling RAID devices.

blkid /dev/sda
/dev/sda: UUID="DELL     ^P(" TYPE="ddf_raid_member"

dmraid -E -r /dev/sda
Do you really want to erase "ddf1" ondisk metadata on /dev/sda? y/n :y
ERROR: ddf1: seeking device "/dev/sda" to 12803086946048
ERROR: writing metadata to /dev/sda, offset 250059349504 sectors, size 0 bytes returned 0
ERROR: erasing ondisk metadata on /dev/sda

booherbg

I have a project that was built 3 years ago, back before Elixir stopped being maintained. As projects have a tendancy to do, it's now time for an update. I've learned a lot in the last 3 years, and am looking forward to adding database migrations to the project. At the same time, if it aint broke don't fix it. I'm keeping Elixir in place, as it works well for our needs. However, Flask-Migrate relies on direct access to an SQLAlchemy object. I'm keeping is simple and going directly to the source: alembic.

The nice thing about this is that I can do the migrations directly on the db, without needing Elixir or any of our app logic.

``` bash
$ alembic init alembic
$ alembic revision -m "adding fields to media: id, filename_uploaded, and tags"
  Generating /data/cliftonlabs/sportsvision/alembic/versions/4b7eedb40b12_adding_fields_to_media_id_filename_.py ... done
```

booherbg

I have a project that's a few years old, but needs some maintenance. Since I last worked on the project, two major things have happened. First, Elixir -- a declarative layer on top of SQLAlchemy -- has not been maintained. Secondly, SQLAlchemy now has its own declarative layer.

Rather than continue using Elixir, I decided to migrate my data models to use the new SQLAlchemy declarative layer, which interestingly enough appear to be relatively compatible with Elixir's philosophy.

The first thing I did was do a direct mapping in my model columns and update the import statements.

# from Elixir import Entity, Field, DateTime, Unicode, Integer, Boolean, setup_all, create_all, session, metadata
from sqlalchemy import Column, DateTime, String, Integer, ForeignKey, Boolean
from sqlalchemy.ext.declarative import declarative_base

booherbg

Interesting examples of meta data by Google

1. Etymology Charts -- when searching for the history of a word, using "etymology" as a keyword in the search, we are presented with a chart that shows interesting information about the word (in this case, 'crass'). For example, I immediately learn that it comes from 'crassus', Latin for 'thick or dense'. I see that it stems from the late 15th century. There is also a graph showing the usage over time from the 1850s, likely based on references in scanned literature (my guess). As a side note, it would be fascinating to see different charts normalized against things like population growth, printing technology improvements, or whether it was driven by a true cultural shift, like an influx in population from one region to another (or, into American cities like Cincinnati), or usage in a popular play (that can reach more people now due to printing press innovations and, lat

booherbg

This is just a general repository for me to keep track of books that I've read over the years.

2018

• The Dosadi Experiment (1977), Frank Herbert
• Snow Crash (1992), Neil Stephenson; This was a doozie. Excellent read.

2017

• Devil in the White City (2003), Erik Larson

booherbg

Google Chrome Pending video status

There's an outstanding bug in Google Chrome where video resources aren't properly closed after download, so the socket limit gets maxed and subsequent videos (and all ajax calls) hang. https://code.google.com/p/chromium/issues/detail?id=234779

I tried, with no success:

• Removing the tag after unload (in an attempt to force the resource to be released)

booherbg

Started having some strange behavior today.

My ruby deploy script (via mina) was crashing on irc notifications.

OpenSSL::SSL::SSLError: SSL_connect returned=1 errno=0 state=unknown state: unknown protocol

Later in the day, a few of our client software scripts could only connect 2-at-a-time (when one would drop, a third would join in its place). When I dug into the guts of it, the python error message was: