It looks like after installing ArcGIS 10.1 SP1, if you have background processing enabled for geoprocessing, you cannot use a VB expression type for the calculate fields tool. Even though VB is still the default expression type.

So if you have a script that calls arcpy.AddField_management() without defining Python as the expression type its not going to run if background processing is enabled...

Sort of a Game

An extension is now available to make writing code a game.

Here is the link

You get "Badges" when you compile code and VS finds stuff.

It could be interesting. I should give it a try when I get back into C# work.

ArcGIS Toolbox: Loading script tool parameter defaults from a config file.

Lets say we create a script tool to run in ArcGIS Desktop and it has 10 parameters. On average the user only needs to change one of those parameters to run the tool 98% of the time.

We make each parameter available because hard coded  parameters can kill your tool 2% of the time. Or maybe more like 10%.... Whatever.

Its best to have all the parameters exposed in the tool dialog and loaded with the defaults your going to use 98% of the time. If these defaults are the same across many tools you don't want to be going into each tool's parameters to set the defaults. You want one location to set all the defaults. That's your config file bro.

So here is an example of one tool's tool validator that loads the defaults from a config file located in the same directory as the toolbox we are running the tool from.


import sys
import os
import ConfigParser
import arcpy

# Finds the path of the toolbox
scriptPath = sys.path[0]
# The config file should be in the same directory as the toolbox
configFile = scriptPath + "\\amp_config.cfg"

class ToolValidator:
  """Class for validating a tool's parameter values and controlling
  the behavior of the tool's dialog."""

  def __init__(self):
    """Setup arcpy and the list of tool parameters."""
    self.params = arcpy.GetParameterInfo()
    #Group the following parameters in the dialog box
    self.params[2].category = "Defaults"
    self.params[3].category = "Defaults"
    self.params[4].category = "Defaults"
    self.params[5].category = "Defaults"
    self.params[6].category = "Defaults"
    self.params[7].category = "Defaults"

  def initializeParameters(self):
    """Refine the properties of a tool's parameters.  This method is
    called when the tool is opened."""
    return

  def updateParameters(self):
    """Modify the values and properties of parameters before internal
    validation is performed.  This method is called whenever a parmater
    has been changed."""
    # If the config file is not found the parameters remain blank.
    if not os.path.exists(configFile):
     self.params[2].value = ""
     self.params[3].value = ""
     self.params[4].value = ""
     self.params[5].value = ""
     self.params[6].value = ""
     self.params[7].value = ""
     return
   
    #If the config file is found it moves on to populate the parameters.
    # Adds defalut values from the config file when the first paramter is filled for the first time.
    if self.params[0].altered and not self.params[0].hasBeenValidated:
      if not self.params[0].value == None:
        # Loads the config file into memory
        config = ConfigParser.SafeConfigParser()
        config.readfp(open(configFile))
       # Gets the values from the config file.
        scratchDir = config.get('environment','scratchDir')
        coverageName = os.path.basename(str(self.params[0].value))
        logFile = scratchDir + "\\" + coverageName + "_log.txt"    
        self.params[3].value = logFile
        self.params[4].value = coverageName
        self.params[6].value = scratchDir + "\\" + coverageName + "_cov.gdb"
        self.params[2].value = scratchDir
        self.params[5].value = config.get('dataConversion','coverageList')
        self.params[7].value = config.get('dataConversion','annotationLayerList')
# If parameter 1 is false then parameter 7 is greyed out.
    if self.params[1].value == False:
      self.params[7].enabled = False
    else:
      self.params[7].enabled = True
    return

  def updateMessages(self):
    """Modify the messages created by internal validation for each tool
    parameter.  This method is called after internal validation."""
    return

ArcGIS Project Management with Production Mapping

ESRI's Production Mapping has a lot of useful tools and functionality. I think its kind of like that drawer in your house where you put things that are really great for something but not necessarily needed for anything.

For me the documentation is rarely clear or concise enough to understand so I just have to tinker with it until I find something that works.

Recently I dove into the Product Library. I am working on a project where the product will be a workflow to create maps. The workflow includes some script tools but the goal is to use as much out of the box functionality as possible. Anyways, the development process requires setting up MXDs, tool boxes, scripts with config files, layout rules, styles, and just about anything else GIS related. It turns out the Product Library is a great way to sync all these things between team members.

I have not really figured the structure yet but after setting up a "Solution"  and a "Product Class" we are able to check in and out everything to a versioned SDE database. If we choose to check everything out to the same location on our respective machines the MXDs, toolboxes and what not stay working.

I'm just saying its neat that's all.

Converting Decimal Degrees to Degrees Minutes Seconds with Python

Thanks to this post at Another GIS Blog I was saved the time of having to figure out how to do this calculation on my own.

And I just the author posted that TODAY, just 8 hours ago. That is crazy.

Thanks Guy.

Tricky Data driven pages

I was having a problem with a data driven pages script where I needed to use the pageRow property to access an attribute. The only examples I found required hard coding the field name but after a little Googleing I found someones suggestion to use:

fieldName = arcpy.GetParameterAsText(1)
mxd.dataDrivenPages.pageRow.getValue(fieldName)


This brings in the field name from the input.

i5 2500k and ASUS P8P67 R3 1155 not boot, not posting, nothing...

I was having problems with my machine, which I just built about 3 months ago. I would hit the power button and the power light on the case would give a quick flash and the CPU and GPU fans would give about one rotation then nothing. If I hit the power button again nothing would happen. So I would turn the PSU off for a few seconds then try again. This would get me the same results 5-10 times until I would get lucky and the machine would start up. Once up the machine worked great. No problems at all for how ever long I had it running. But after turning it off it was back to the same dance to get it going again.

Everything in the build was 3 months old with the exception of my OCZ 600w power supply and my hard drive.. After trying about everything I could find on the google machine I decided it must be the power supply.

After picking up a new PSU and testing it I found that was the issue. I have had the new PSU in for a few days now and done a bunch of cold starts and no problems...

I am disappointed that my OCZ psu died. It was a nice PSU and only 3 years old. I always kept it clean too...

Oh well.

This post is an object

This post has been instantiated and so its an object now. I think.

I have been focusing on learning ArcObjects on .NET via C# the last couple weeks and the last few days I finally feel like I am starting to get it. I think the key has been grasping the Object Oriented approach. These articles are what really got be moving with that.They are pretty generic and quick to get through, which helps me.
http://www.developer.com/lang/article.php/3628381/The-Object-Oriented-Thought-Process-Index-Page.htm

As for C# in particular this series has helped me get a grasp on the basics.
http://msdn.microsoft.com/en-us/beginner/bb308730.aspx

Other than that I want to start posting something useful so...

I found that the ArcObject IQuerDef2 only allows "advanced" SQL queries on SDE databases... or at least not on file geodatabases. For example, the following works on an SDE but not File geodatabase.


 IFeatureWorkspace featureWorkspace = workspace;
            IQueryDef2 queryDef = (IQueryDef2)featureWorkspace.CreateQueryDef();
            queryDef.Tables = inTable;
            queryDef.PrefixClause = "TOP (29)"; 
            queryDef.SubFields = field + ", COUNT(*)";
            queryDef.PostfixClause = "GROUP BY " + field + " ORDER BY COUNT(*) DESC, " + field; 
            queryDef.WhereClause = field + "!= ''"; 

I never found this documented anywhere but after testing a ton of queries on both sources its clear.

Maybe someday there will be an IQuerDefX that will support it. Or maybe it will come down to the File GDB format since the SQL probably gets thrown to the database anyways.

Map Symbology

So lets say you have a fancy map (ArcMap MXD) and you spend lots of time tweaking symobology. You can save that symbology one by one or you can get your self the Export Map Styles button and do it with two clicks.

This is sort of hidden now on Arc 10 but its still there and it can save you a ridiculous amount of time.

More details here: http://blogs.esri.com/Support/blogs/mappingcenter/archive/2010/09/28/Symbols-and-Styles-in-ArcGIS-10.0.aspx

Getting more of what someone paid for.

I am starting to really like Visual Studio 2010. My new favorite thing is historical clipboard.

Ctrl+Shift+V will cycle through the clipboard's history allowing you to paste things you copied a while ago. Nice when you are hacking stuff together.

You get what [someone] pays for.

I am starting to use Visual Studio (Professional) to write python. The first thing I had trouble with was missing the function in Pyscripter that highlights all occurrences of a selected word . A little trip to the google got me an extension that does this. I got it here.  Woot.

Data Flow Diagram... Ok

I have to do my first "official" data flow diagram. I don't really know what I am doing yet but after rummaging around for the an approach to take I found that Google Docs drawings has flow chart functionality. It looks like it will do everything I need.

Scratch Names

I was always wishing there was an easy way to create a unique scratch name for outputs of geoprocessing scripts. Today I found the simple solution http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#/CreateScratchName/000v0000001z000000/

The Create Scratch Name code does just that so if you are making a temporary copy of  features or a temporary buffer or whatever you can slap this line in and have it make names for you. Its great for someone like me that has to run their code 70 times before finishing it.

Why did it take me so long to learn about this?

I just came across a little tool in Windows 7 that might come in handy. Its a recorder that records your mouse actions and takes screen shots along the way. When your done it puts it in a little html and zips it up. You can even edit the html in Word to clean it up or use it for "how to's" and what not. Seems like this could be a very useful little guy.

To run just hit the Windows key and type psr.exe.

Too much of a good thing

This is random but I wanted to write it down.

We had 55 attendees at our wedding. Our wedding events included a welcome bbq on Friday night, wedding and reception on Saturday and general hanging out on Sunday and we had 30 people staying on site with us Friday-Sunday morning. Given these factors we thought we would need a lot of alcohol so we acquired 3 kegs several buckets of Margarita and several 3 liter boxes of wine. I don't know what we were thinking.The kegs were available to our guests from about 3pm on Friday until Monday morning around 9am. Despite everyones best efforts only 1 Coors lite keg was fully consumed. The other Coors Lite keg was only lightly dented and the Mirror Pond keg was about 3/4 drank. We also had a fair amount of wine left over.

The moral of the story is unless your party consists of lots of college kids, Don't buy so much booze.

Unique IDentification

I don't know how I missed this, but the Global ID (GUID) field in ArcGIS gives you a hands free way to add unique ids to features. The Add Global IDs tool will create the field and populate it with the ID. The GlobalID will stick with the features as you edit or join or export or clip or whatever. This will allow you to trace back and maintain relationships.

I will start using this in everything. No more staring at the attribute table to see what I can combine to make a Unique ID.

Metadata is for the birds

Today I was faced with some metadata that ended up in HTML but needed to go back into a geodatabase as FGDC metadata. Luckily I found this link that outlined out to use a USGS tool to convert the HTML to a text file that could be imported. No manual ctrl+c and ctrl+v required.

Many tools for many jobs

Creating multi page tiffs or pdfs from multiple images is super easy with ImageMagick. After installing it you run it from the command line. "d:\whereveryouwant>convert image1.tif image2.jpg image3.tif mymergedimage.pdf" This simple command appends your multiple images, they can be anywhere, and makes a multi page tif or pdf. There are other options to the command if you want to look it up.

Hellsyeah Gmail style message origination in Outlook 2010

I got bored and Googled Office 2010 tweaks. To my joy I found that you can get Gmail style message consolidation in Outlook 2010. This means when there is a thread of emails Outlook will chain them together instead of keeping a copy of each individually. It was my main complaint about Outlook and its solved. WOOT!

Read the link for this and other 2010 tweaks.


Office 2010 tweaks

Breaking up large files

I have been running into tables that are too big for Excel and needed an easy way to break them into smaller chunks. Unix to the rescue on that. If you have a Unix box like Ubuntu or emulator like Cygwin all you need to do is run the command line code split option filename prefix. Split has two options -l (by line number) or -b (size in bytes, I think).

So I had the giant table in a csv, I ran "split -l 500000 d:\folder\folder\bigfile.csv smallerfile" and my big csv was broken up into 500k line csvs. From there I just opened them in a text editor and copied the header line into each new file.