Interested in advanced Excel training? Peltier Tech is planning sessions for Summer and Fall 2018. If interested, fill out this survey: http://goo.gl/kDx5za
@TweetingDuck I’ve always wondered how much I’d benefit from going to one of these. I still feel like I don’t know as much as I should regarding excel.
@Mat'sMug Value2 can never return a Date or Currency type:
> The only difference between this property and the Value property is that the Value2 property doesn't use the Currency and Date data types. You can return values formatted with these data types as floating-point numbers by using the Double data type.
so I got the keys in a neutral format. That's good.
Now I need to transform them for 32-bit and 64-bit install.
In the Inno Setup, we can either say HKCU, or HKCU32 or HKCU64
does it make sense to generate 3 registry tables for each?
The annoying thing is that examining the keys, only Record and TypeLib lack their virtualized entries. I have CLSID and Interface in both places. Because those came from the DLL, I have a hunch that those are on the account of the Any CPU setting. The TLB file, OTOH is specific to one or other platform. Yet, my entries for the TLB and the Record aren't in virtualized path despite being 32-bit.
interestingly, TLB has a key to indicate the bitness. I wonder what'd happen if it had keys in both....
@ThunderFrame you have both 32-bit and 64-bit VBA hosts?
@this Yep. I have 32-bit Office running on 64-bit OS, and a VM running 64-bit Office on 64-bit OS.... Soon to have a VM with 32-bit Office running on 32-bit Windows...
@Mat'sMug We know that Worksheets.Item will either succeed or fail, depending upon whether the argument is valid, but that's not the reason that Item returns a Variant... It's because the argument can be an array, and if it is, then the return value is an array of Worksheets, otherwise it's just a Worksheet.
@this I didn't - it was more than a regasm, IIUC, because I would have needed to copy over the 64-bit file(s) too?
so i guess the TypeLib should NOT be going to the virtualized place....
but I can't be sure it works
registering the 64-bit type library does not appear to affect the 32-bit RD, at least.
so it looks to me that I only need to generate platform-specific keys for the subkeys to Interface and CLSID and platform-agnostic keys for the Record and TypeLib
i think that's a something bit different. Proof by contradiction demonstrates that this assertion is false but it's not an induction that every other assertions are false.
Law of non-contradiction is simply an assertion that an A cannot be not A {at same time and space}
and that's how we derive the other principles of logic such as identity A is A, excluded middle if A is not B then B is not A and few more. All those derives from that idea A cannot be not A
math builds on it, so it cannot be illogical. At least I don't think it can be.
Just for reference, any logical system with a finite set of rules capable of describing arithmetics cannot be proven to be consistent from within itself.
On the other hand, first order logic, the one computers are build on, is consistent.
@Vogel612 That sounds a lot like YouTube saying "we're going to allow you to post your video, but we disagree with it and we're going to point out why" without actually saying it out loud. /late to the party
"We're going to allow you to post your video, but because we get a lot of flak for recommending shitty conspiracy and indoctrination videos like yours because they produce engagement we're going to add a symbolic disclaimer to cover our backsides"
@M.Doerner I assume we're talking about Gödel's incompleteness theorem. However, I always understood even the first order logic are themselves accepted a priori because there is no way to prove the law of non-contradiction (and therefore its derivatives) as true.
> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=h1) Report > Merging [#3836](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/443a039b294800b97b6061747ebb98b61cc1a160?src=pr&el=desc) will **not change** coverage. > The diff coverage is `n/a`.
> The last PR introduces the idea of a registry writer which at the moment is the Inno Setup, the idea being that it will generate a `.iss` file that can be then `#incldue`d into the main `.iss` script. The code is incomplete because there needs to be function to handle the check. Note that the registry outputs will be written for both 32-bit and 64-bit, and TLBs will be exported for both bitness. Because the DLL is `Any CPU`, we don't need 32/64 bit version of the DLL; only the TLB.
> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=h1) Report > Merging [#3836](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/443a039b294800b97b6061747ebb98b61cc1a160?src=pr&el=desc) will **not change** coverage. > The diff coverage is `n/a`.
> The last PR introduces the idea of a registry writer which at the moment is the Inno Setup, the idea being that it will generate a `.iss` file that can be then `#incldue`d into the main `.iss` script. The code is incomplete because there needs to be function to handle the check. Note that the registry outputs will be written for both 32-bit and 64-bit, and TLBs will be exported for both bitness. Because the DLL is `Any CPU`, we don't need 32/64 bit version of the DLL; only the TLB.
[Currently awaiting moderation on this post. Wanna to bet I'll get a reply a decade later?](https://blogs.msdn.microsoft.com/larryosterman/2006/01/11/what-registry-entries-are-needed-to-register-a-com-object/#comment-185665)
Larry, even a decade later, this is still helpful. There is a small detail that’s puzzling to me. When I inspect the output of regasm.exe, it does not always write everything to the same side of the registry. For example, on an assembly with “Any CPU”, registy entries will be written for both 32-bit and 64-bit branch of the tree but only for CLSID and Interface. For Ty…
"if [...] the minimum required parameters are the system id and the passwords (specified in XML syntax and streamed in, or specified in the configuration file)."
@Hosch250 - if you get a job at MS, please, please, I implore you, find a what to make them stop changing application interfaces for no apparent reason and without a way of setting them back to the way they were.
/HATES all the extra line spacing that comes as a default in Outlook
There are valid reasons for forcing the extra line spacing in between Inbox folders in Outlook on a desktop machine without any touch input? I'd be willing to pay to sign the NDA to find out what they are.
@Hosch250 Doubt it, the interface in 2010 was far from ugly, and it had a very normal amount of vertical spacing. Granted, fat fingers on a phone wouldn't have been very effective at working it, but not everybody is using the app on a phone.
If the changed it as they did for 2016 but gave an option to reduce the spacing (for example when the system doesn't detect touch input), it would be much more usable.
@this but there is no good default when you're trying to develop one codebase for a 24" 1920x1200 monitor w/o touch and a 5.5" phone with 1080x720 or even full HD resolution.
Oh, a lot of people that benefit from a higher-contrast theme (not full-high-contrast) don't think they need it, so they don't switch and struggle to enjoy the site.
@FreeMan UserVoice is but one channel; insider feedback is another... mostly used by MVP's. UV priorities don't necessarily reflect the real-world priorities... Python for Excel (4K votes+) is a good example :)
Actually, I think python support for Office would make sense. Not because of data science, but because it is primarily reference counted, in contrast to c# and vb.Net.
#TIL Formulas>FormulaAuditing>TracePrecedents will have the precedent arrow be a thicker line if it refers to more than 1 cell. It'll also have an outline around the multi-column/row range.
VBA Rubberducking
VBA Rubberducking