![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Columbus Day... ain't it?
Oct. 14th, 2002 09:29 pmSpoke to two CPAs today. The difference between them is like night and day, making it easy to decide who to go with. I had my suspicions, anyway.
The paper chase continues. In a way, it's a little like programming. You get to a point where you feel you're 90% finished, and then realize you have another 90% to go.
I haven't done a lot of digging, but I cannot - off the top of my head - imagine what purpose there might be to blowing up a nightclub on Bali. One report I read said that the media in Australia - which is where a significant number of victims came from - is downplaying the terrorist angle.
* * * I sent the translation, as agreed, to client T by deadline. When I called a few minutes later to confirm receipt, I got their answering machine. Now understand, I used to work there, and I know the president of the company, and she has a very simple rule for telephone protocol: the receptionist is to answer all calls by the third ring, and it will be a cold, cold day in hell before she'll inflict a voicemail system on her customers.
Indeed, I think the answering machine was put in place only because some people at NASA were under the impression that the company was a round-the-clock shop, seeing as how they would call at all hours and be very disappointed at not having anyone to talk to at, say, 2 am.
As someone who hates corporate voicemail systems with a passion, this trait is one of many that I like about her.
Anyway, I guess what I was trying to say was that the answering machine being on during business hours on a Monday could only mean one thing: the place was closed for the Columbus Day holiday.
* * * I did stop by the store briefly while Galina got Sasha from the vet's. (The poor thing came back with a bandage on her forepaw where the intravenous drip had been installed, and completely woozy.) We went through the same thing we do on every federal holiday when UPS and FedEx deliver and pick up: having to tell people who wanted to send mail that we couldn't sell postage because of the holiday. Fortunately, most people understand, but the whole idea kind of rankles me. Ah, well.
* * * After a day of chasing paper, I sat down to revise a program I started writing a couple of years ago. It's a command-line program that takes Cyrillic text files as input, analyzes their content, determines which coding is used, and then converts the coding if such a conversion is specified.
I call the poor thing the Cyrillic Code Conversion Program (cccp for short, and yes, the name is deliberate), and it does its thing among the Alternative, KOI8-R, and CP1251 codings. Increasingly, however, I'm finding it useful to convert text to Unicode, and started to wonder what it would take in terms of new code.
This goal meshes nicely with my proposed talk at the ATA Conference in Atlanta ("Finding your way through the Cyrillic swamp" is, if memory serves, the subtitle of my presentation).
When I saved a sample file from Word 97 as "Unicode text," I was surprised to find the first two bytes in the file were 0xff and 0xfe, respectively. It turns out that the integer value 0xfeff (or U+FEFF in Unicodese) is a zero-width no-break space that's used as a "byte order" mark to distinguish files created on little-endian Intel machines (where the value 0xabcd is stored as 0xcd followed by 0xab, i.e., with the bytes of the value swapped).
I've given up the mod for the evening, having run across a roadblock in writing an unsigned integer such as 0xfeff to stdout. It turns out, after consulting with fellow LJer![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
teferi, that perhaps fwrite() might be a better choice for the task at hand than my attempt to use fprintf(). I'll pursue this again tomorrow, with the new information.
And that's because I've been up since early (got up at 5:30 am and started to check the translation... good thing, too... as I made the deadline by a whisker), so the day is starting to nail me. I'm going to take Sasha upstairs, too. She's been sitting in the middle of my office for most of the day, moving barely enough to lap up some water from time to time. Letting her go for a walk is out of the question, as she can barely walk (I suspect it's the phenobarbitol tablet I put down her throat a while back).
Cheers...
The paper chase continues. In a way, it's a little like programming. You get to a point where you feel you're 90% finished, and then realize you have another 90% to go.
I haven't done a lot of digging, but I cannot - off the top of my head - imagine what purpose there might be to blowing up a nightclub on Bali. One report I read said that the media in Australia - which is where a significant number of victims came from - is downplaying the terrorist angle.
Indeed, I think the answering machine was put in place only because some people at NASA were under the impression that the company was a round-the-clock shop, seeing as how they would call at all hours and be very disappointed at not having anyone to talk to at, say, 2 am.
As someone who hates corporate voicemail systems with a passion, this trait is one of many that I like about her.
Anyway, I guess what I was trying to say was that the answering machine being on during business hours on a Monday could only mean one thing: the place was closed for the Columbus Day holiday.
I call the poor thing the Cyrillic Code Conversion Program (cccp for short, and yes, the name is deliberate), and it does its thing among the Alternative, KOI8-R, and CP1251 codings. Increasingly, however, I'm finding it useful to convert text to Unicode, and started to wonder what it would take in terms of new code.
This goal meshes nicely with my proposed talk at the ATA Conference in Atlanta ("Finding your way through the Cyrillic swamp" is, if memory serves, the subtitle of my presentation).
When I saved a sample file from Word 97 as "Unicode text," I was surprised to find the first two bytes in the file were 0xff and 0xfe, respectively. It turns out that the integer value 0xfeff (or U+FEFF in Unicodese) is a zero-width no-break space that's used as a "byte order" mark to distinguish files created on little-endian Intel machines (where the value 0xabcd is stored as 0xcd followed by 0xab, i.e., with the bytes of the value swapped).
I've given up the mod for the evening, having run across a roadblock in writing an unsigned integer such as 0xfeff to stdout. It turns out, after consulting with fellow LJer
teferi, that perhaps fwrite() might be a better choice for the task at hand than my attempt to use fprintf(). I'll pursue this again tomorrow, with the new information.And that's because I've been up since early (got up at 5:30 am and started to check the translation... good thing, too... as I made the deadline by a whisker), so the day is starting to nail me. I'm going to take Sasha upstairs, too. She's been sitting in the middle of my office for most of the day, moving barely enough to lap up some water from time to time. Letting her go for a walk is out of the question, as she can barely walk (I suspect it's the phenobarbitol tablet I put down her throat a while back).
Cheers...
