从C＃读取Excel文件(Reading an Excel File From C#)

 y轴没有限制（除了数字类型的javascript限制外） 
 
 但是，您可能看到的可能是9007199254740992以上的精度损失 - 这不是巧合的17位数（请参阅什么是JavaScript的最高整数值，数字可以达到而不会丢失精度？以及相关问题 
 
 Chart.js使用值范围来确定如何缩放条形。 它通过从最大值中减去min并使用差值来实现。 
 
 现在，如果所有值都相同，Chart.js会添加一个小数字（0.5）来产生差异并继续使用此差异。 
 
 如果添加0.5实际产生差异，那么一切都很好 - 但是如果结果高于9007199254740992，它将大部分被精度限制吞噬，给出0差异并弄乱Chart.js的y轴计算。 
 
 当数字高于9007199254740992并且它们的差异不够大时会发生同样的问题（它们基本上将相同的数字向上舍入 - >它们被视为一组相等的值 - > 0.5被添加 - >它主要被吞噬了精度限制） 
 
 这是一个快速的价值示例，以及为什么它会起作用/不起作用 
 
// works because there is a (small) difference and the numbers are < 9007199254740992
var a = [9007199254740991, 9007199254740992];

// won't work because there is no difference and the numbers are > 9007199254740992 - 0.5
var b = [9007199254740992, 9007199254740992];

// won't work because there is no difference and the numbers are = 9007199254740992 - 0.5
var c = [9007199254740991.5, 9007199254740991.5];

// works because there is no difference and the numbers are < 9007199254740992 - 0.5
var d = [9007199254740991.4, 9007199254740991.4];

// works because there is a significant difference even though the numbers are > 9007199254740992
var e = [12345678901234567890, 12345678901434567891];

// works because there is a significant difference even though the numbers are > 9007199254740992
var f = [0, 12345678901434567891];

// won't work because there is only a small difference and the numbers are > 9007199254740992
var g = [9007199254740992, 9007199254740993];
 
 简而言之，您的条形图不会渲染，因为您有一组相同（或没有显着差异）的值，并且是9007199254740992 - 0.5或更高。 
 
 您可以添加一个虚拟值0来强制它渲染（或者可能添加另一个虚拟系列的0值），确保存在显着差异 - 或者如果所有值都在17位数范围内并且非常彼此接近，只绘制偏移量即表示你有1701和1705 ...图1和5并在工具提示中输入170作为值前缀。 

There is no limit on the y-axis (except possibly the javascript limit on the number type)
 
However, what you are possibly seeing might be a LOSS of precision above 9007199254740992 - which not so coincidentally is a 17 digit number (see What is JavaScript's highest integer value that a Number can go to without losing precision? and the associated problems
 
Chart.js uses the value range to figure out how to scale the bars. It does this by subtracting the min from the max and using the difference. 
 
Now, if all values are same, Chart.js adds a small number (0.5) to make a difference and proceeds with this difference.
 
All is well and good if adding this 0.5 actually produces a difference - however if the result is above 9007199254740992, it will mostly get swallowed up by the precision limit giving a 0 difference and messing up Chart.js's y axis calculations.
 
The same issue happens when the numbers are above 9007199254740992 and their difference is not large enough (they basically round up the same number -> they are treated as a set of equal values -> 0.5 is added -> it mostly get swallowed up by the precision limit)
 
Here's a quick sample of values and why it will work / not work
 
// works because there is a (small) difference and the numbers are < 9007199254740992
var a = [9007199254740991, 9007199254740992];

// won't work because there is no difference and the numbers are > 9007199254740992 - 0.5
var b = [9007199254740992, 9007199254740992];

// won't work because there is no difference and the numbers are = 9007199254740992 - 0.5
var c = [9007199254740991.5, 9007199254740991.5];

// works because there is no difference and the numbers are < 9007199254740992 - 0.5
var d = [9007199254740991.4, 9007199254740991.4];

// works because there is a significant difference even though the numbers are > 9007199254740992
var e = [12345678901234567890, 12345678901434567891];

// works because there is a significant difference even though the numbers are > 9007199254740992
var f = [0, 12345678901434567891];

// won't work because there is only a small difference and the numbers are > 9007199254740992
var g = [9007199254740992, 9007199254740993];
 
In short, your bar is not rendering because you have a set of values that are same (or do not significantly differ) AND are 9007199254740992 - 0.5 or above.
 
You could just add a dummy value of 0 to force it to render (or possibly add another dummy series of 0 values) by making sure there is a significant difference - or if all your values are going to be in the 17 digit range and very close to each other, just plot the offset i.e. say you have 1701 and 1705... plot 1 and 5 and put in the 170 as a value prefix in the tooltip.