blocks|key|195505|text|推测：|type|unstyled|depth|inlineStyleRanges|entityRanges|data|195506|以下是似乎正在发生的事情。|195507|一旦对象被创建，knitr就会非常明智地缓存这些对象。然后，每当检测到它们被更改时，它都会更新它们的缓存值。|offset|length|style|BOLD|195508|但是，data.table绕过了R的常规按值复制赋值和替换机制，并使用:=运算符，而不是=、<<-或<-。因此，knitr没有接收到DT已被DT[,+c:=5]更改的信号。|CODE|195509|解决方案：|195510|只需将此块添加到代码中希望重新缓存DT的当前值的任何位置。它不会消耗任何内存或时间(因为除了引用之外，DT+<-+DT不会复制任何其他内容)，但它确实会有效地向knitr发送一个(假的)信号，表明DT已经更新：|195511|```{r,+cache=TRUE,+echo=FALSE}|code-block|syntax|javascript|195512|DT+<-+DT|195513|195514|195515|示例文档的工作版本：|195516|通过运行您的文档的以下编辑版本来检查它是否正常工作：|195517|```{r}|195518|库(data.table)|195519|Data.Table+Markdown
========================================================
Suppose+we+make+a+`data.table`+in+**R+Markdown**
```{r,+cache=TRUE}|195520|DT+=+data.table(a+=+rnorm(10))|195521|Then+add+a+column+using+`:=`
```{r,+cache=TRUE}|195522|DT，c:=5|195523|```{r,+cache=TRUE,+echo=FALSE}|195524|DT+<-+DT|195525|Then+we+display+that+in+a+non-cached+block
```{r,+cache=FALSE}|195526|DT|195527|The+first+time+you+run+this,+the+above+will+show+a+`c`+column.+
The+second,+third,+and+nth+times,+it+will+as+well.|195528|entityMap^0|0|0|8|5|0|3|A|1K|5|Z|2|18|1|1A|3|1E|2|1U|2|1Y|A|0|0|H|2|1F|8|2Q|2|28|5|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0^^$0|@$1|2|3|4|5|6|7|1R|8|@]|9|@]|A|$]]|$1|B|3|C|5|6|7|1S|8|@]|9|@]|A|$]]|$1|D|3|E|5|6|7|1T|8|@$F|1U|G|1V|H|I]]|9|@]|A|$]]|$1|J|3|K|5|6|7|1W|8|@$F|1X|G|1Y|H|I]|$F|1Z|G|20|H|I]|$F|21|G|22|H|L]|$F|23|G|24|H|L]|$F|25|G|26|H|L]|$F|27|G|28|H|L]|$F|29|G|2A|H|L]|$F|2B|G|2C|H|L]]|9|@]|A|$]]|$1|M|3|N|5|6|7|2D|8|@]|9|@]|A|$]]|$1|O|3|P|5|6|7|2E|8|@$F|2F|G|2G|H|L]|$F|2H|G|2I|H|L]|$F|2J|G|2K|H|L]|$F|2L|G|2M|H|I]]|9|@]|A|$]]|$1|Q|3|R|5|S|7|2N|8|@]|9|@]|A|$T|U]]|$1|V|3|W|5|6|7|2O|8|@]|9|@]|A|$]]|$1|X|3|-4|5|S|7|2P|8|@]|9|@]|A|$T|U]]|$1|Y|3|-4|5|6|7|2Q|8|@]|9|@]|A|$]]|$1|Z|3|10|5|6|7|2R|8|@]|9|@]|A|$]]|$1|11|3|12|5|6|7|2S|8|@]|9|@]|A|$]]|$1|13|3|14|5|S|7|2T|8|@]|9|@]|A|$T|U]]|$1|15|3|16|5|6|7|2U|8|@]|9|@]|A|$]]|$1|17|3|18|5|S|7|2V|8|@]|9|@]|A|$T|U]]|$1|19|3|1A|5|6|7|2W|8|@]|9|@]|A|$]]|$1|1B|3|1C|5|S|7|2X|8|@]|9|@]|A|$T|U]]|$1|1D|3|1E|5|6|7|2Y|8|@]|9|@]|A|$]]|$1|1F|3|1G|5|S|7|2Z|8|@]|9|@]|A|$T|U]]|$1|1H|3|1I|5|6|7|30|8|@]|9|@]|A|$]]|$1|1J|3|1K|5|S|7|31|8|@]|9|@]|A|$T|U]]|$1|1L|3|1M|5|6|7|32|8|@]|9|@]|A|$]]|$1|1N|3|1O|5|S|7|33|8|@]|9|@]|A|$T|U]]|$1|1P|3|-4|5|6|7|34|8|@]|9|@]|A|$]]]|1Q|$]]

<h2>Speculation:</h2>

Here is what appears to be going on.

knitr quite sensibly caches objects as as soon as they are created. It then updates their cached value whenever it detects that they have been altered. 

data.table, though, bypasses R's normal copy-by-value assignment and replacement mechanisms, and uses a <code>:=</code> operator rather than a <code>=</code>, <code>&lt;&lt;-</code>, or <code>&lt;-</code>. As a result knitr isn't picking up the signals that <code>DT</code> has been changed by <code>DT[, c:=5]</code>.

<h2>Solution:</h2>

Just add this block to your code wherever you'd like the current value of <code>DT</code> to be re-cached. It won't cost you anything memory or time-wise (since nothing except a reference is copied by <code>DT &lt;- DT</code>) but it does effectively send a (fake) signal to knitr that <code>DT</code> has been updated:

<pre><code>```{r, cache=TRUE, echo=FALSE}
DT &lt;- DT 
```
</code></pre>

<hr>

<h2>Working version of example doc:</h2>

Check that it works by running this edited version of your doc:

<pre><code>```{r}
library(data.table)
```
Data.Table Markdown
========================================================
Suppose we make a `data.table` in **R Markdown**
```{r, cache=TRUE}
DT = data.table(a = rnorm(10))
```

Then add a column using `:=`
```{r, cache=TRUE}
DT[, c:=5] 
```

```{r, cache=TRUE, echo=FALSE}
DT &lt;- DT 
```

Then we display that in a non-cached block
```{r, cache=FALSE}
DT
```
The first time you run this, the above will show a `c` column. 
The second, third, and nth times, it will as well.
</code></pre>

blocks|key|3033816|text|正如Josh+O‘’Brien在答案下面的第四条评论中指出的那样，我已经添加了一个新的块选项cache.vars来处理这种非常特殊的情况。在第二个缓存块中，我们可以指定cache.vars='DT'，以便knitr保存DT的副本。|type|unstyled|depth|inlineStyleRanges|offset|length|style|CODE|entityRanges|data|3033817|```{r}|code-block|syntax|javascript|3033818|库(data.table)|3033819|Data.Table+Markdown
========================================================
Suppose+we+make+a+`data.table`+in+**R+Markdown**
```{r,+cache=TRUE}|3033820|DT+=+data.table(a+=+rnorm(10))|3033821|Then+add+a+column+using+`:=`
```{r,+cache=TRUE,+cache.vars='DT'}|3033822|DT，c:=5|3033823|Then+we+display+that+in+a+non-cached+block
```{r,+cache=FALSE}|3033824|DT|3033825|3033826|无论您编译文档多少次，输出都是这样的：|3033827|​|3033828|📷|atomic|3033829|3033830|entityMap|0|IMAGE|mutability|IMMUTABLE|imageUrl|https://ask.qcloudimg.com/http-save/yehe-900000/9928fff4229f27eaa90b50e2f8a5c0eb.png|imageAlt^0|1A|A|2C|F|2U|5|31|2|0|0|0|0|0|0|0|0|0|0|0|0|0|1|0|0|0^^$0|@$1|2|3|4|5|6|7|1G|8|@$9|1H|A|1I|B|C]|$9|1J|A|1K|B|C]|$9|1L|A|1M|B|C]|$9|1N|A|1O|B|C]]|D|@]|E|$]]|$1|F|3|G|5|H|7|1P|8|@]|D|@]|E|$I|J]]|$1|K|3|L|5|6|7|1Q|8|@]|D|@]|E|$]]|$1|M|3|N|5|H|7|1R|8|@]|D|@]|E|$I|J]]|$1|O|3|P|5|6|7|1S|8|@]|D|@]|E|$]]|$1|Q|3|R|5|H|7|1T|8|@]|D|@]|E|$I|J]]|$1|S|3|T|5|6|7|1U|8|@]|D|@]|E|$]]|$1|U|3|V|5|H|7|1V|8|@]|D|@]|E|$I|J]]|$1|W|3|X|5|6|7|1W|8|@]|D|@]|E|$]]|$1|Y|3|-4|5|H|7|1X|8|@]|D|@]|E|$I|J]]|$1|Z|3|10|5|6|7|1Y|8|@]|D|@]|E|$]]|$1|11|3|12|5|6|7|1Z|8|@]|D|@]|E|$]]|$1|13|3|14|5|15|7|20|8|@]|D|@$9|21|A|22|1|23]]|E|$]]|$1|16|3|12|5|6|7|24|8|@]|D|@]|E|$]]|$1|17|3|-4|5|6|7|25|8|@]|D|@]|E|$]]]|18|$19|$5|1A|1B|1C|E|$1D|1E|1F|-4]]]]

As indicated in the fourth comment under the answer by Josh O'Brien, I have added a new chunk option <code>cache.vars</code> to handle this very special case. In the second cached chunk, we can specify <code>cache.vars='DT'</code> so that <code>knitr</code> will save a copy of <code>DT</code>.

<pre><code>```{r}
library(data.table)
```
Data.Table Markdown
========================================================
Suppose we make a `data.table` in **R Markdown**
```{r, cache=TRUE}
DT = data.table(a = rnorm(10))
```
Then add a column using `:=`
```{r, cache=TRUE, cache.vars='DT'}
DT[, c:=5] 
```
Then we display that in a non-cached block
```{r, cache=FALSE}
DT
```
</code></pre>

The output is like this no matter how many times you compile the document:

<img src="https://i.stack.imgur.com/mk1Pi.png" alt="knitr works with data.table now">

This is related in spirit to <a href="https://stackoverflow.com/q/15267018/1900520">this</a> question, but must be different in mechanism.

If you try to cache a <code>knitr</code> chunk that contains a <code>data.table</code> <code>:=</code> assignement then it acts as though that chunk has not been run, and later chunks do not see the affect of the <code>:=</code>.

Any idea why this is? How does <code>knitr</code> detect objects have updated, and what is <code>data.table</code> doing that confuses it?

It appears you can work around this by doing <code>DT = DT[, LHS:=RHS]</code>.

Example:

<pre><code>```{r}
library(data.table)
```
Data.Table Markdown
========================================================
Suppose we make a `data.table` in **R Markdown**
```{r, cache=TRUE}
DT = data.table(a = rnorm(10))
```
Then add a column using `:=`
```{r, cache=TRUE}
DT[, c:=5] 
```
Then we display that in a non-cached block
```{r, cache=FALSE}
DT
```
The first time you run this, the above will show a `c` column, 
from the second time onwards it will not.
</code></pre>

Output on second run

<img src="https://i.stack.imgur.com/1VTSN.gif" alt="knitr output">

why does knitr caching fail for data.table `:=`?

这在精神上与问题有关，但在机制上必须不同。如果您尝试缓存包含data.table :=分配的knitr块，那么它的行为就像该块尚未运行一样，并且以后的块看不到:=的影响。知道为什么会这样吗？knitr是如何检测对象已经更新的，data.table做了什么让它困惑的事情？看起来你可以通过执行DT = DT[, LHS:=...

问为什么data.table `:=`的caching缓存失败？
EN

社区

活动

资源

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问为什么data.table `:=`的caching缓存失败？EN

社区

活动

资源

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问为什么data.table `:=`的caching缓存失败？
EN