## QUERY leaves (Select, Where, Contains, Limit) that works, except when the cell is empty

I have a QUERY formula of Google sheets that works pretty much:
= query (data, "SELECT A WHERE B contains & # 39;" & A5 & "LIMIT 10", 1)

Essentially, the query does, for the given data range, it selects a cell in column A where column B contains some text, and I limit the output list of values ​​returned to just 10 rows. Everything works well.

Where I encounter a problem, is when the "& A5 &" above refers to an empty cell, I always get a list of returned (and bad) values ​​when really, if an empty cell is referenced, the output should be nothing at all. So, if "& A5 &" refers to an empty cell, I do not want ANY output value returned, that is, empty results.

I did a quick Google search on several different forums to find possible solutions. Apparently, "is not zero" could be a solution? Or <> & # 39; But I do not know how to integrate in the syntax.

My apologies, I am not an expert in sheets / queries / syntax and I would appreciate any advice / suggestions.

Thank you

## Indexer Limit Option Question RE

Can you explain how the limit of indexing works? If I set the indexing option to send only 30,000 links per day to the indexing service but GSA creates links of 40,000, what happens to them? 10,000 others? Are they sent the next day or ..?

## macos – Internet sharing from Macbook Pro has a limit of 5 devices?

I'm trying to share the internet connection (wifi) from a Macbook Pro 15 inch 2019 on OSX version 10.14.6 (18G1012) using a USB Ethernet connector and the built-in WiFi chipset of the. device as my "router". The first 5 devices all work (2 iPhones, a PS4, an iPad, a Macbook Pro 2015 on 10.15), but as soon as I connect any other device that I have (Xbox One, Nintendo Switch, Macbook Pro 2013), they will all do so in a predictable way with a generic error message that can be summarized as "can not connect" and will ask for the Wi-Fi password again. I share the connection of my USB Ethernet adapter as:

If I switch the shared Wi-Fi connection and start everyone, if I choose a device that has failed to connect before the others, it will work until 5 devices are connected, and # 6 and beyond appliance will still fail. Even though he previously knew the network! If I disable Wi-Fi from a currently connected device, a faulty device will connect. It is pretty clear that there is a limit of 5 devices on the sharing laptop.

I've done some research and it seems like others have no problem connecting more than 5 devices to a Macbook's wifi, so what do I do? wrong? Do not they connect simultaneously as I am? I'm not afraid of the terminal or gather more information, but I do not even know where to start. The sharing screen is not terribly talkative.

## topology gt. geometric – limit connection sum of Ganatra-Pardon-Shende

In section 3.4 of https://arxiv.org/pdf/1809.03427.pdf, Ganatra-Pardon-Shende defines the sum of the connections to the boundaries of two exact conical lagrangians in a Liouville domain. In particular, in Figure 10, they explain what happens to the legendary at the border in terms of frontal projection of the legendary.

Question: At the level of the Legendrians, can we rather describe this surgery in terms of Lagrangian projection of the Legendrians?

Is it just Polterovich's surgery at the point of transverse intersection determined by Reeb's agreement? If so, why? For example, is there a simple image describing equivalence in dimension 3?

## ubuntu – Nginx reaches the limit of worker connection

I manage a Nginx server that continues to lose a lot of connections occasionally, after inspecting the logs, I can see that the `worker_connections` the limit is reached, eg

``````2019/12/06 08:37:09 (alert) 14517#14517: 25000 worker_connections are not enough
``````

I can see from my measurements that Nginx starts to drop connections after reaching 20k active connections despite the fact that there are 32 (self) workers, each of whom should be able to handle 25000 connections (if my understanding is correct).

What I have tried:

• Increase in the worker connection limit
• Turned `multi_accept` sure
• Epoll activated – I could see that the majority of the load was handled by 2-3 work processes and that others were virtually unused, although that does not really seem to help the situation because the PID displayed in the error log is almost always the same process.

I am really puzzled with what else to try, if anyone has any suggestions on what might be causing it, it would be greatly appreciated.

Other relevant information:

• 32 processors, 64 GB of memory
• Ubuntu v18.04.3
• Connections are mandated upstream
• Nginx is averaging about 400 RPS

## plugins – Quiz time! Is there a way to limit publisher editing permissions so that it can only edit posts from specific contributors?

We are forming a new team of writers, with a new publisher. I want the new editor to edit the messages created by the new team. But I do not want the publisher to be able to edit existing publications, or pages, most of which have been created by administrators.

Is there a way to do that, or a plugin that we could buy that would give us that capability? Thank you for your help! :]

## privacy – What is the justification for the lightning network path length limit (20 jumps)?

A Lightning Network sender can choose a path with up to 20 intermediates (BOLT4).

Tor is another network that uses onion routing to ensure anonymity. However, in Tor, all paths have 3 intermediaries and the longest paths are explicitly discouraged (source 1, source 2). The sources point out that the attacks are mainly based on the compromise of the first and last nodes. Increasing the number of nodes between the two does not improve anonymity, but increases the latency.

Do similar arguments apply to Lightning (perhaps for runs of slightly longer lengths but less than 20 hops)? Where does the number 20 come from? Would it be a good idea to decrease it?

## linear algebra – Spectral space limit of a simple symmetric matrix

I have a seemingly innocent linear algebra problem that I can not solve and on which I hope you would like to give a glimpse. Here is the description: Let $$mathbf {a} = (a_1, a_2, dots, a_d) ^ {T}$$ to be a positive probability vector, $$c is$$ $$Green mathbf {a} Green_1 = 1$$ and $$a_i> 0$$ for everyone $$i$$. Leave matrix $$A$$ to be defined as follows: $$A = textrm {diag} ( mathbf {a}) – mathbf {a} mathbf {a} ^ {T}$$ or $$textrm {diag} ( mathbf {a})$$ means the diagonal matrix with the $$i$$the diagonal being $$a_i$$. It's simple to show that $$mathbf {1} _d$$, the all-a dimension vector $$d$$, is a clean vector of $$A$$ of own value $$0$$. And Gershgorin's circle theorem also shows that all $$A$$Eigenvalues ​​of are greater than or equal to $$0$$. My question is:

What is the smallest eigenvalue of $$A$$ it's not nil?

I did the calculation when $$d = 3$$ and realized that there may not be a simple analytical formula and that it was therefore greatly appreciated to have a nice lower limit.

Thank you so much!

## security – In Baggy Bounds Checking, why is an error generated when accessing more than half the size of the slot the end limit of an allocated memory?

The second note of the MIT Computer Systems Security Class (6.858) contains this excerpt (slightly modified).

``````char *p = malloc(44);
char *q = p + 60;
char *r = q + 16;
// r is now at an offset of 60+16=76 from
// p. This means that r is (76-64)=12 bytes
// beyond the end of p. This is more than
// half a slot away, so baggy bounds will
// raise an error.
``````

I understand that Baggy Bounds Checking makes it possible to check at a lower cost if a pointer to an allocated memory is or is not within the limits of this memory. This prevents many forms of attacks involving the replacement of arbitrary parts of memory. What I do not understand, it is this part of the extract above.

This is more than half a time slot, too big limits will create an error.

Why should an error be generated when accessing memory more than half the size of the slot from the final limit of an allocated memory?

## applications – Limit the number of installed applications

My kids both have an Android tablet. With parental control, I can limit the type of applications to PEGI 7 for example, which works well.

But then they continue to install applications until the device is full, and then no longer work properly.

What I'm looking for is an option to limit the number of applications installed at the same time. So I can set this number, for example, 30 or 50. If they want to install a new one, they must first delete one.

Googling on this subject, I find a lot of things about time management, but I already have an application to fix that. It is therefore only necessary to define the maximum number of installed applications.

I hope someone has an idea of ​​how to achieve this, either through a setting in Android or through a third-party application.

Thank you!