Christmas Price Index

Nov 16, 2021

In my recent sleep-deprived state, I started to think about the Christmas song The Twelve Days of Christmas. Being a cumulative song, I've always found it funny that, technically speaking, the gift-giver ends up giving the true love 364 gifts. I got looking at the Wikipedia page for the song, and lo and behold, there's an actual economic indicator (albeit a tongue-in-cheek one) that PNC Wealth Management maintains about these gifts. It's called the Christmas Price Index, and gets used as a commodity price index. More amusing than the index itself are some of the criticisms of it that are listed in the Wikipedia article:

First, the index does not clearly define the products that comprise each of the twelve gifts. For example, the price for the eight "maids a-milking" only includes the cost of eight laborers at Federal minimum wage, while milking also requires at least a milk cow, goat, or other such animal, which is an additional cost.

Even better is this note on the index for 2020:

The 2020 index did not include nine Ladies Dancing, ten Lords-A-Leaping, eleven Pipers Piping, or twelve Drummers Drumming due to COVID-19 restrictions on live performances.

Brilliant.

CSV Parsing Woes

Nov 13, 2021

An occasional annoyance of my job is having to deal with poorly constructed data. One recent instance of this came through a collection of CSV files. In these files, certain free-form text fields sometimes included either non-escaped double quotes or an embedded newline where there shouldn't be one. Shortened examples of each are shown below:

"Samsung","ABC-12345","2.5 TB SAS 2.5" hard drive","Released","2018-06-01"
"Lenovo","DEF-88776 
PQR-66554","Mechanical chassis","Released","2020-02-22"

The first line above has an embedded double quote character which has not been escaped. The second line showcases a rogue newline character.

Parsing these problematic cases in Python gets real tricky, and the native csv module doesn't have great malformed data handling support. While thinking about how to handle these situations, it occurred to me that I could use the way the file was constructed to my advantage. These files are output by, what is to me, a black box. Under the hood it's undoubtedly a database query, the results of which are then sent into a CSV format. As a byproduct, each file has a consistent format where each field is quoted, and fields are separated by a comma. I can use the "," string (double quote, comma, double quote) as my separator, looking for the fields I expect:

previous_chunk = []
with open(infile, 'r', encoding='utf8') as csvfile:
    with open(f"{infile.stem}-clean.csv", 'w', encoding='utf8', newline='') as outfile:
        writer = csv.writer(outfile, quoting=csv.QUOTE_ALL)

        for line in csvfile.readlines():
            line = line.rstrip()  # Trim the trailing newline

            pieces = line.split('","')  # Split on our separator
            pieces[0] = re.sub(r'^"', '', pieces[0])  # Remove the first double quote
            pieces[-1] = re.sub(r'"$', '', pieces[-1])  # Remove the last double quote

            # If we don't have the number of columns we expect, merge
            if(len(pieces) != expected_columns):
                previous_chunk = merge_chunks(previous_chunk, pieces)
                if(len(previous_chunk) == expected_columns):
                    writer.writerow(previous_chunk)
                    previous_chunk = []
                elif(len(previous_chunk) > expected_columns):
                    print(f"ERROR: Overran column count! Expected {expected_columns}, Found "
                          f"{len(previous_chunk)}")
            else:
                writer.writerow(pieces)

The merge_chunks method is very simple:

def merge_chunks(a, b):
"""
Merges lists a and b. The content of the first element of list b will be appended
to the content of the last element of list a. The result will be returned.
"""
    temp = []
    temp.extend(a)

    if(a):
        temp[-1] = f"{a[-1]} {b[0]}"
        temp.extend(b[1:])
    else:
        temp.extend(b)

    return temp

I believe the only way this could potentially break is if the content, for some reason, contained the "," separator somewhere in a data field. Given the types of data fields I'm working with, this is highly unlikely. Even if it does occur, I can use the format of some of the fields to make best guesses as to where the actual dividers are (e.g. the trailing elements on each line are typically always date stamps).

This is obviously not a general solution, but it sometimes pays to step away from the built-in parsing capability in a language and roll your own scheme.

I recently had to change the URLs of some of my REST endpoints in a project at work. In so doing, I started receiving reports of users who weren't seeing the data they expected from some endpoints. My redirect seemed simple enough:

location ~* ^/rest/(report/.*)$ {
    return 302 https://$host/rest/project/$1;
}

So, I'm sending URLs like https://myhostname/rest/report/1/ to https://myhostname/rest/project/report/1/. The redirect worked, but the problem was that any additional query string that was included in the URL got dropped.

For example, https://myhostname/rest/report/1/?query=somevalue would result in https://myhostname/rest/project/report/1/. The fix for this is easy, but is something I didn't realize you had to pay attention to:

location ~* ^/rest/(report/.*)$ {
    return 302 https://$host/rest/project/$1$is_args$args;
}

The $is_args directive will return a ? if there are arguments, or an empty string otherwise. Similarly, $args will return any arguments that happened to be passed. Including these variables ensures that any query parameters get passed along with the redirect.

Here's another terrific Tom Scott video, this time showcasing a shooting range in Switzerland where you shoot over a busy road. The technology behind the electronic targets is pretty neat. One more reason why Switzerland is one of the best places on Earth!

I didn't know this, but Judy Sheindlin (of Judge Judy fame) was interviewed back in 1993 on CBS' 60 Minutes. This interview is what essentially launched her career as a television judge. It's fascinating that her personality on television is essentially the same one she used in New York's family court system. It turns out that she's pretty much herself!

Barred Owl

Sep 30, 2021

I've longed to photograph (and even just see) an owl in the wild for a long time, but they are oh-so-elusive birds. We do hear barred owls behind our house regularly, but I never see them. However, just before sundown tonight, one started calling out. I ventured into the woods to find it and, fortunately, I was able to!

Barred owl

Office Chairs

Sep 23, 2021

I've been fortunate during this pandemic to be able to work from home the entire time. It has only recently occurred to me, however, that my current setup is lousy when it comes to comfort. My desk is the same one I've had since my college days, and the chair I use is a hand-me-down from my wife (which is a step up from the crappy cheap one I used to own). Why am I putting up with this?

My ideal desk would be a standing model, if for no other reason than to provide some height adjustability; my current desk is a fixed height, and it's rather low down in my opinion. The ideal chair, however, is harder for me to pin down. Should it have a head rest? Arm rests? Mesh or padded backing?

If you're reading this, what kind of chair do you use? Where can one go to try out office chairs before you buy? I don't recall seeing these in furniture stores, and office supply stores (as I recall) only carry the cheap stuff.

This follow-up video from Technology Connections on dishwasher soap is fascinating (the video is in response to a previous one on dishwasher detergent packs). It's over 45 minutes long, so you'll need to make time for it, but I give it a high recommendation. The previous one is also well worth a watch, though he provides a very quick overview of it in this video.

For over a decade now, I have logged my daily weight using Libra Weight Manager, a great little Android app. I last wrote about this in December of 2018, so it's high time for an update. Shown below is the latest graph (click to expand):

As you can see, I've gained weight substantially. In fact, I'm now the heaviest I've ever been! The graph starts heading up in 2018, which happened to be the year our first daughter was born, though I started gaining well before she arrived. At the beginning of the pandemic in 2020, my weight started to come back down, but has since done an about-face and is increasing again.

Stress likely corresponds (to a degree) with my weight drops. The lowest point on the graph is just before I got married, and the big dip at the beginning of 2020 is pandemic related. My nightly weakness is ice cream; I have some almost every night. This, and an increase in the snacking I do in the afternoon (thanks to working from home) are definite contributing factors. Eating at home has had benefits, however, as I'm eating better meals than I used to for lunch. Prior to working from home, I ate out nearly every day.

I need to turn things around again and lose some weight. I'd certainly feel better, both mentally and physically.

Back in November of 2019, I wrote about some deformable LED lights that I picked up on Amazon for our garage. This weekend, one of them died, after only about 21 months of occasional use. I have since picked up a newer variant that claims to have a 5-year warranty. The happy part is that I got a pack of 2 of these for much cheaper than I paid for one previously.

It's disappointing to me that I didn't even get 2 years out of those bulbs, despite their having a "2-year warranty." Lots of Amazon reviews for the old model point out that the warranty is bogus; no one is available at this "company" to take your information and replace the product. Given how many clones of these there are, I'm guessing they're all pretty much bogus Chinese listings. I am unable to find any UL-listed variants of these lights, which is also disappointing. I have read about some of these melting, especially the ones with plastic bodies. Both of the ones I own have aluminum bodies, which should help with temperature control, but I'd feel a lot better if the bulbs were certified in some way.

Ultimately, I'd love to install some of the consistently-well-reviewed Barrina LED fixtures in my garage, but that would involve some rewiring work that I'm not real keen on doing.

Projection Clock

Aug 23, 2021
Projection clock

My eyesight is pretty bad, which makes reading the time on the clock next to the bed a challenge at night. Since 2016, my wife and I have used a projection clock / radio combination, but it had a number of drawbacks:

  • The display uses bright blue numbers, which is hard to look at in the dark
  • The display is either too dim to read in the daytime (and just right at night), or it's just right in the daytime and too bright at night
  • The time from the projection feature is too small for my poor vision
  • The projection feature is also too dim to see in all but the darkest rooms
  • Our cats chewed up the tiny antenna on the radio, rendering it useless (not that we used it much anyways)

Frustrated with all of these features, I picked up this projection alarm clock from Amazon. It's terrific:

  • This clock uses red numbers, which are way easier to read at night
  • The primary display isn't too bright or dim (and you can control the brightness across 4 levels)
  • The projected time is larger than my previous clock
  • The brightness of the projected time is controllable; I use the brightest setting, which makes the numbers quite readable on the ceiling
  • There's no radio to fool with (or antenna for cats to chew)
  • It has a USB port to slow-charge your phone, which is nice but not something I plan to use

At only $25 (I got it on sale for $20), it's a nice improvement to our bedroom. The only drawback I can think of so far is that the clock is ridiculously light, making it easy to slide around on the bedside table.

Yellow-billed Cuckoo

Aug 10, 2021
Yellow-billed cuckoo

This morning I got really lucky and photographed the 37th different bird species I've seen in our backyard: a yellow-billed cuckoo. These birds are apparently notorious for being hard to see, though they are frequently heard due to their distinctive calls. A number of people in a North Carolina birding forum that I post to have heard them in the wild, but have never seen one. Lucky catch!

I spotted it out of the kitchen window this morning, had an idea as to what it might be, and grabbed my camera. I sat outside on our back deck for 5 or 10 minutes before it showed itself again. Though the photo isn't the greatest, I'm happy that I was able to get a snapshot of it.

One of the great thrills of birding is checking off birds on your life list. This "lifer" for me was a fun one to get, and gives me the bug even more to find (and photograph) new birds!

We use AG-Grid at work for several of our projects. Earlier this week, I ran into an interesting issue in some code being used to load data into the grid. The call was very simple:

agGrid.simpleHttpRequest({url: theURLToLoad })
    .then(function(data) {
        gridOptions.api.setRowData(data);
    });

This asynchronous call was failing and no error was being thrown (as we typically do elsewhere in our code). In looking around, I couldn't determine where the simpleHttpRequest call was defined. The AG-Grid documentation, which is generally pretty good, had no mention of it, save for its use in a few examples. After half an hour of digging, I decided to actually poke around in the AG-Grid source code. There I found the function's definition:

function simpleHttpRequest(params) {
    return new _utils__WEBPACK_IMPORTED_MODULE_0__["Promise"](function (resolve) {
        var httpRequest = new XMLHttpRequest();
        httpRequest.open('GET', params.url);
        httpRequest.send();
        httpRequest.onreadystatechange = function () {
            if (httpRequest.readyState === 4 && httpRequest.status === 200) {
                resolve(JSON.parse(httpRequest.responseText));
            }
        };
    });
}

All this function is an incredibly light-weight wrapper around XMLHttpRequest. It provides absolutely no support for error handling, and is missing all the other hooks one would need to take full control over the request and its response. It frustrates me when packages do stuff like this. If you're going to show how to fetch remote data for your package, use the vanilla JavaScript features that are industry standard, not some custom wrapper.

Brave Mobile Browser

Aug 3, 2021

Over the past few weeks, I've switched my primary mobile browser (in Android) from Firefox to Brave (thanks for the suggestion, Kip!). So far, I'm very impressed with the app. The built-in ad blocker is impressively accurate, and the browser is incredibly fast. In fact, it feels faster to me than Chrome does (and it's night and day compared to Firefox). There are a few minor UI annoyances here and there, but nothing that ever gets in the way. I've also been impressed with the updates the development team pushes out (the browser only seems to get better with time). I recommend it.

Lately I've been thinking about image post-processing a fair amount. This is partly due to my shooting more photos, which subsequently need to be processed before I share them (I shoot in RAW). Post-processing is an area where I feel I have room to improve as a photographer. Most of the room for improvement comes down to the time I'm willing to put into the post-processing step. Often, I'm eager to share my photos, and will do only a small amount of editing to minimize the time necessary for that step. Investing time in this step, however, can often result in better photos. It's a deep rabbit hole, however, and I occasionally find myself asking "how far is too far?" when it comes to photo adjustments.

The tutorial videos I've seen online run the spectrum of possibilities. Some photographers treat their images very conservatively, making only the bare minimum changes to bring out the best of the photo. Others seemingly "run amok" with changes: from removing parts of the scene, to wildly adjusting the color of the lighting (often to make it appear that the photo was taken at a different time of day). Personally, I think I tend to lean towards the former thinking: adjusting only what's necessary.

I spent some time this afternoon making an alternate edit of a recent photo I took of a barn swallow at Lake Lynn in Raleigh, NC. Take a look at the following two images (click to enlarge):

The first image was my original "quick pass" at editing: global adjustments only. The second image was done with a little more care: I adjusted specific sections of the photo individually, spending more time in the process. I also adjusted the crop of the second image (the first's crop was a little off, in my opinion).

This particular bird was backlit in the photograph, making the side towards the camera much darker (as can be seen in the first image). I brightened that up considerably for the second image. Are the adjustments in the second image too much? Who's to say, really? It's an interesting problem to think about. Where is the line between reality and creative freedom?

My ultimate goal is to produce the best possible pictures I can. Much of that process begins with taking better photos to begin with, as post-processing can only help so much. Taking better photos requires practice which, happily, I'm getting more of with each passing day.