xyzio

Archive for December 2019

PlugIn Board – An AVRISP breakout board for prototyping with the AVR microcontroller

leave a comment »

In the past I would connect AVR microcontrollers to the AVRISP programmer by using individual wires to connect the 6×2 IDC connector to the protoboard.  There was also the hassle of wiring up a 5v supply.  So I designed the PlugIn Board to quickly connect the AVRISP programmer to AVR microcontrollers on a protoboard.

Layout

Looking at the PCB Board Layout below, the PlugIn board takes the AVRISP 2×3-pin IDC connector (ISP1) and maps to a 0.1-inch in-line header that plugs into a protoboard.  In addition it can power the AVR. Unregulated power can be fed through a 2.1mm DC jack or screw-down terminal connector(J1) which is connected to a 7805 regulator(IC2) to provide 5V to the AVR and other circuits.

A few other features in the PlugIn board is it has the recommended 10kOhm pull-up resistor(R1) for reset and the pinout matches the ATMEGA32 programming and power pins.

PCB Board Layout

Future Improvements

If I were to do another iteration of the PlugIn Board I’d add the recommended decoupling circuitry for the ADC with a separate ground plane as outlined in the Atmel AVR design considerations app notes.

While not recommended due to the long signal path, I’d also include a header for an external clock.  It would allow the user to swap out crystal oscillators or use an external clock source without fiddling around on the protoboard.

PCB Board Files

The PCB is designed using Eagle.  The board and schematics are available on my BitBucket at https://bitbucket.org/xyzio/circuits/src/master/PlugIn%20Board/

Final Product

I had the board fabbed at OSH Park.  It works well and greatly simplifies programming AVR microcontrollers.

Written by M Kapoor

December 25, 2019 at 7:22 pm

Posted in microcontrollers, Project

Tagged with ,

Generating Charts with Python and matplotlib as Base64 images for embedding in HTML webpages

leave a comment »

This is code used to create the charts for my ‘Compare Expense Ratios‘ microsite. It uses matplotlib to generate a chart which is converted to a Base64 to be embedded directly into a webpage. This way there is no need to save an intermediate image.

#Inputs are a dictionary of lists that contains points to plot (points), the title, and the x and y-axis labels.
def mathPlotLib(points, title, xlabel, ylabel):
   import matplotlib
   matplotlib.use('Agg')
   from matplotlib import pyplot
   import base64
   import cStringIO
   import natsort

   #The two funds are the keys to the dicitonary
   fundOrder = []
   for fund in natsort.natsorted(points.keys()):
      #Plot the points and keep track of the order they are added for the legend
      pyplot.plot(points[fund])
      fundOrder.append(fund)

   # Add the legend, title, and x,y labels.
   pyplot.legend(fundOrder)
   pyplot.title(title)
   pyplot.xlabel(xlabel)
   pyplot.ylabel(ylabel)

   # Convert the image as png as a byte-string object
   my_stringIObytes = cStringIO.StringIO()
   pyplot.savefig(my_stringIObytes, format='png')

   # Seek to the beginning of the file and encode it as Base64
   my_stringIObytes.seek(0)
   b64png = base64.b64encode(my_stringIObytes.read())

   # Add the html wrapper to embed it as base64
   html = '<img src="image/png;base64,' + b64png + '" />'

   # Clean up
   my_stringIObytes.close()
   pyplot.clf()
   pyplot.close()

   return html

 

Written by M Kapoor

December 24, 2019 at 8:04 pm

Posted in Programming, python

Tagged with , ,

Automate build and load of Hugo sites to Amazon S3 using Rclone with Python

leave a comment »

This was the build and load flow for my Hugo site before I gave up on Hugo and moved back to WordPress. The site was generated using Hugo and pushed to Amazon s3 using rclone.

The code assumes the hugo and rclone executables, and the hugo root directory are in the same folder.

In this example the hugo directory is called codepearls.xyzio.com.


import os, shutil

# s3 auth info
s3SecretAccessKey = 's3_secret_key'
s3AccessKeyId = 's3_access_key'

# Hugo folder and bucket name.  sitepath is path to the hugo public folder.
path = 'codepearls.xyzio.com'
sitepath = os.path.join(path, 'public')

#Remove the files from the previous build by deleting the public folder
if os.path.exists(sitepath):
   shutil.rmtree(sitepath)

# Run hugo from root directory
cmd = 'hugo.exe -s ' + path
os.system(cmd)

# Run rclone from root directory to sync /public to s3 and the appropriate args to encrypt the files and enable public read
cmd = 'rclone.exe sync ' + sitepath + ' s3:' + path
cmd += ' -v --s3-secret-access-key ' + s3SecretAccessKey
cmd += ' --s3-access-key-id ' + s3AccessKeyId
cmd += ' --s3-acl public-read'
cmd += ' --s3-server-side-encryption AES256'
cmd += ' --s3-provider AWS'
cmd += ' --s3-region us-west-2'

os.system(cmd)

Written by M Kapoor

December 24, 2019 at 3:26 am

Posted in python

Tagged with , ,

Split an Excel file’s tabs into CSV with C#

leave a comment »

Split an Excel file into csv, with each tab going into a separate csv file, in the background using C#.


using System.IO;
using Microsoft.Office.Interop.Excel;

Microsoft.Office.Interop.Excel.Application app = new Microsoft.Office.Interop.Excel.Application();

//Run in background
app.DisplayAlerts = false;

//Open excel file
Workbook wb = app.Workbooks.Open("path_to_excel.xls");

//Iterate through sheets in the excel file
foreach (Worksheet sheet in wb.Worksheets)
{
    string sheetName = sheet.Name;
    
    //Output file name
    string outputFilepath = sheetName + ".csv";
    
    //Save the sheet as CSV
    sheet.SaveAs(outputFilepath, XlFileFormat.xlCSVWindows);
}

//Clean-up
wb.Close();
app.Quit();


Written by M Kapoor

December 24, 2019 at 3:19 am

Posted in c#, Programming

Tagged with ,

Open a zip archive and iterate through its files with Python

leave a comment »

Open a file compressed as .zip and iterate through its files line by line with Python.


import zipfile

#Path to the zip file
zipfilepath = 'path_to_zip_file'

#Read in zip file
zip = zipfile.ZipFile(zipfilepath)

#Iterate through files in zip file
for zipfilename in zip.filelist:
    
    #Read contents of the file
    filecontents = zip.read(zipfilename)
    
    #Break up contents into list and process
    for line in filecontents.replace('\r\n', '\n').split('\n'):
        print line

Written by M Kapoor

December 24, 2019 at 3:09 am

Posted in Programming

Tagged with ,

Open and print a .gz.bz2 file with Python

leave a comment »

Open and print a file that is compressed with gzip and then with bzip2, i.e a gz.bz2 file, with Python.


import sys
import bz2
import gzip
from cStringIO import StringIO

# .gz.bz2 File is given as commandline argument
filename = sys.argv[1]

gzbzfilename = filename

#Read in the bz2 data
o = open(gzbzfilename, 'rb')
gzbzdata = o.read()

#Decompress the bz2 data
gzdata = bz2.decompress(gzbzdata)

#Next, gunzip the gzip file and read out the file pointer
f = gzip.GzipFile(fileobj=StringIO(gzdata))
file_content = f.read()
f.close()

#Print out the file contents
print file_content

Written by M Kapoor

December 24, 2019 at 3:04 am

Posted in Programming

Tagged with , , ,

Use ffmpeg To Create a Timelapse From Pictures With Music

leave a comment »

Create a timelapse movie from images using ffmpeg with background music.

Images must be sequentialy numbered starting with the value given for the start_number argument.

Command:

ffmpeg -framerate 12 -start_number 15 -f image2 -y -i "G%7d.jpg" -i music.mp3 -c:v libx264 -preset fast -c:a mp3 -shortest -movflags +faststart "output_filename.mp4"

 

Options Breakdown:

Option Description
framerate 12 Number of images to show per second
start_number 15 Start number of the first image
-i “G%7d.jpg” ffmpeg formatted file name
-i music.mp3 Background music filename
-shortest Truncate video to shortest of timelapse or music
-movflags +faststart Put file index at beginning to speed up online streaming

Written by M Kapoor

December 23, 2019 at 11:54 pm

Posted in Programming

Tagged with ,